<?xml version="1.0" encoding="UTF-8"?>
<rdf:RDF xmlns="http://purl.org/rss/1.0/"
 xmlns:dc="http://purl.org/dc/elements/1.1/"
 xmlns:dcterms="http://purl.org/dc/terms/"
 xmlns:cc="http://web.resource.org/cc/"
 xmlns:prism="http://prismstandard.org/namespaces/basic/2.0/"
 xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
 xmlns:admin="http://webns.net/mvcb/"
 xmlns:content="http://purl.org/rss/1.0/modules/content/">
    <channel rdf:about="https://www.mdpi.com/rss/journal/automation">
		<title>Automation</title>
		<description>Latest open access articles published in Automation at https://www.mdpi.com/journal/automation</description>
		<link>https://www.mdpi.com/journal/automation</link>
		<admin:generatorAgent rdf:resource="https://www.mdpi.com/journal/automation"/>
		<admin:errorReportsTo rdf:resource="mailto:support@mdpi.com"/>
		<dc:publisher>MDPI</dc:publisher>
		<dc:language>en</dc:language>
		<dc:rights>Creative Commons Attribution (CC-BY)</dc:rights>
						<prism:copyright>MDPI</prism:copyright>
		<prism:rightsAgent>support@mdpi.com</prism:rightsAgent>
        						<items>
			<rdf:Seq>
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/193" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/192" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/191" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/190" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/189" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/188" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/187" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/186" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/185" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/184" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/183" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/182" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/181" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/180" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/179" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/178" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/177" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/176" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/175" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/174" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/173" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/172" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/171" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/6/170" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/169" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/168" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/167" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/165" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/166" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/164" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/163" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/162" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/160" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/161" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/159" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/158" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/157" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/156" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/155" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/154" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/153" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/152" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/151" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/150" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/149" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/148" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/147" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/146" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/145" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/144" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/143" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/142" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/141" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/140" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/139" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/138" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/137" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/136" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/135" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/134" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/5/133" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/132" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/131" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/130" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/129" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/128" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/127" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/126" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/125" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/124" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/123" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/122" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/121" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/120" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/119" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/118" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/117" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/116" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/115" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/114" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/113" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/112" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/111" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/110" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/109" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/108" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/107" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/106" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/105" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/104" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/103" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/102" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/101" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/4/100" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/3/99" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/3/98" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/3/97" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/3/96" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/3/95" />
            				<rdf:li rdf:resource="https://www.mdpi.com/2673-4052/7/3/94" />
                    	</rdf:Seq>
		</items>
				<cc:license rdf:resource="https://creativecommons.org/licenses/by/4.0/" />
	</channel>

        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/193">

	<title>Automation, Vol. 7, Pages 193: Adaptive PID-Fuzzy Control for Trajectory Tracking of Par-4 Delta Parallel Robot</title>
	<link>https://www.mdpi.com/2673-4052/7/6/193</link>
	<description>The Par-4 Delta parallel robot is widely utilized in industrial applications requiring high operational speed and acceleration. However, its control design is severely challenged by strong nonlinearities, cross-coupling dynamics, and structural uncertainty. To overcome these limitations, this study proposes an adaptive trajectory-tracking control framework for the Par-4 Delta parallel robot by integrating Type-1 and Interval Type-2 Fuzzy Logic Controllers (T2FLC) with a Proportional&amp;amp;ndash;Integral&amp;amp;ndash;Derivative (PID) architecture. The proposed scheme is designed to simultaneously enhance position tracking precision, minimize control energy consumption, and ensure robust disturbance rejection. The proposed controllers are evaluated on a spiral trajectory-tracking benchmark using Root Mean Square Error (RMSE) and Root Mean Square Control Effort (RMSCE) as performance metrics. The adaptive Type-2 fuzzy-PID hybrid (APID-T2FLC) achieves the lowest tracking error and reduces control effort compared with standard PID and Type-1 fuzzy-hybrid baselines, demonstrating its suitability for high-precision control of the Par-4 Delta parallel robot. Robustness is further validated under the effects of disturbance and noise, where stable performance has been maintained under implementation of the APID-T2FLC controller rather than other implemented controller types.</description>
	<pubDate>2026-11-18</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 193: Adaptive PID-Fuzzy Control for Trajectory Tracking of Par-4 Delta Parallel Robot</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/193">doi: 10.3390/automation7060193</a></p>
	<p>Authors:
		Shaymaa M. Mahdi
		Ahmed I. Abdulkareem
		Amjad J. Humaidi
		Safanah M. Raafat
		</p>
	<p>The Par-4 Delta parallel robot is widely utilized in industrial applications requiring high operational speed and acceleration. However, its control design is severely challenged by strong nonlinearities, cross-coupling dynamics, and structural uncertainty. To overcome these limitations, this study proposes an adaptive trajectory-tracking control framework for the Par-4 Delta parallel robot by integrating Type-1 and Interval Type-2 Fuzzy Logic Controllers (T2FLC) with a Proportional&amp;amp;ndash;Integral&amp;amp;ndash;Derivative (PID) architecture. The proposed scheme is designed to simultaneously enhance position tracking precision, minimize control energy consumption, and ensure robust disturbance rejection. The proposed controllers are evaluated on a spiral trajectory-tracking benchmark using Root Mean Square Error (RMSE) and Root Mean Square Control Effort (RMSCE) as performance metrics. The adaptive Type-2 fuzzy-PID hybrid (APID-T2FLC) achieves the lowest tracking error and reduces control effort compared with standard PID and Type-1 fuzzy-hybrid baselines, demonstrating its suitability for high-precision control of the Par-4 Delta parallel robot. Robustness is further validated under the effects of disturbance and noise, where stable performance has been maintained under implementation of the APID-T2FLC controller rather than other implemented controller types.</p>
	]]></content:encoded>

	<dc:title>Adaptive PID-Fuzzy Control for Trajectory Tracking of Par-4 Delta Parallel Robot</dc:title>
			<dc:creator>Shaymaa M. Mahdi</dc:creator>
			<dc:creator>Ahmed I. Abdulkareem</dc:creator>
			<dc:creator>Amjad J. Humaidi</dc:creator>
			<dc:creator>Safanah M. Raafat</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060193</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-11-18</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-11-18</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>193</prism:startingPage>
		<prism:doi>10.3390/automation7060193</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/193</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/192">

	<title>Automation, Vol. 7, Pages 192: Energy-Aware Cooperative Multi-Agent Deep Reinforcement Learning for Sustainable and Precision Trajectory Tracking of an Industrial Robotic Arm: An SDG-Aligned Approach</title>
	<link>https://www.mdpi.com/2673-4052/7/6/192</link>
	<description>Control software that lowers actuator effort without degrading precision can be deployed on installed industrial robots without new hardware. This paper formulates trajectory tracking of a 5-degree-of-freedom (DOF) Mitsubishi RV-2AJ arm as a fully cooperative multi-agent problem, in which each joint is driven by a twin delayed deep deterministic policy gradient (TD3) agent under centralized training and decentralized execution (CTDE). Unlike MADDPG, MATD3 and MAPPO, the proposed CTDE-TD3 combines (i) a single twin centralized critic, shared by all joint agents and conditioned on all joint states and torques, which removes the non-stationarity faced by independent agents while each actor uses only three local measurements at run time; (ii) a shared reward with an inter-joint torque-covariance term, whose exact decomposition yields the admissible weight condition μ ≤ 2λ; and (iii) a dimensionless weight normalization. In simulation on three 3D reference trajectories over 10 training seeds, the controller reduced the tracking root-mean-square error (RMSE) by 54.5% and 37.5% and the squared-torque effort by 33.1% and 21.0% relative to centralized and independent-agent TD3, respectively. It was not dominated by MADDPG, MATD3, MAPPO or a tuned proportional–integral–derivative (PID) controller on unseen trajectories. Ablations and a capacity-matched centralized baseline show that the gains stem from the cooperative structure rather than from network size. Because the squared-torque index reflects control effort and copper loss rather than total electrical energy, electricity and CO2 savings are reported only as scenario estimates pending hardware validation.</description>
	<pubDate>2026-11-18</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 192: Energy-Aware Cooperative Multi-Agent Deep Reinforcement Learning for Sustainable and Precision Trajectory Tracking of an Industrial Robotic Arm: An SDG-Aligned Approach</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/192">doi: 10.3390/automation7060192</a></p>
	<p>Authors:
		Zied Ben Hazem
		</p>
	<p>Control software that lowers actuator effort without degrading precision can be deployed on installed industrial robots without new hardware. This paper formulates trajectory tracking of a 5-degree-of-freedom (DOF) Mitsubishi RV-2AJ arm as a fully cooperative multi-agent problem, in which each joint is driven by a twin delayed deep deterministic policy gradient (TD3) agent under centralized training and decentralized execution (CTDE). Unlike MADDPG, MATD3 and MAPPO, the proposed CTDE-TD3 combines (i) a single twin centralized critic, shared by all joint agents and conditioned on all joint states and torques, which removes the non-stationarity faced by independent agents while each actor uses only three local measurements at run time; (ii) a shared reward with an inter-joint torque-covariance term, whose exact decomposition yields the admissible weight condition μ ≤ 2λ; and (iii) a dimensionless weight normalization. In simulation on three 3D reference trajectories over 10 training seeds, the controller reduced the tracking root-mean-square error (RMSE) by 54.5% and 37.5% and the squared-torque effort by 33.1% and 21.0% relative to centralized and independent-agent TD3, respectively. It was not dominated by MADDPG, MATD3, MAPPO or a tuned proportional–integral–derivative (PID) controller on unseen trajectories. Ablations and a capacity-matched centralized baseline show that the gains stem from the cooperative structure rather than from network size. Because the squared-torque index reflects control effort and copper loss rather than total electrical energy, electricity and CO2 savings are reported only as scenario estimates pending hardware validation.</p>
	]]></content:encoded>

	<dc:title>Energy-Aware Cooperative Multi-Agent Deep Reinforcement Learning for Sustainable and Precision Trajectory Tracking of an Industrial Robotic Arm: An SDG-Aligned Approach</dc:title>
			<dc:creator>Zied Ben Hazem</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060192</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-11-18</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-11-18</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>192</prism:startingPage>
		<prism:doi>10.3390/automation7060192</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/192</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/191">

	<title>Automation, Vol. 7, Pages 191: Multi-Objective AGV Task Allocation and Path Planning: Evaluation of Deterministic and Metaheuristic Algorithms</title>
	<link>https://www.mdpi.com/2673-4052/7/6/191</link>
	<description>Automated guided vehicles (AGVs) enable material transport in automated manufacturing facilities, where fleet efficiency depends on coordinating task allocation and route generation while balancing completion time, energy consumption, and travel distance. This study develops a Python(version: 3.14.6)-based multi-objective evaluation framework for offline AGV path planning and task assignment in grid-based warehouse environments with known, static obstacles. Particle swarm optimization (PSO) and the gravitational search algorithm (GSA) are evaluated against deterministic A* search across warehouse scenarios with different layouts and obstacle configurations. Each planner is integrated into a common greedy assignment procedure that processes tasks according to priority and earliest-start ordering. For each task, candidate AGVs are evaluated using planned routes from their current positions to the pickup location and subsequently to the delivery location. Assignment decisions minimize a weighted cost combining estimated completion time, route energy consumption, and travel distance. Fleet-level performance is assessed using makespan, defined as the maximum completion time across AGVs, together with aggregate energy consumption and total travel distance. Energy estimates account for motion-related costs, including acceleration and turning, enabling comparisons beyond geometric path length alone. The benchmark results indicate that A* provides consistently competitive performance, particularly in energy consumption, whereas PSO achieves favorable makespan&amp;amp;ndash;distance trade-offs in selected scenarios. As a future extension toward online path planning, the A* algorithm, which supports real-time path generation, is integrated with NSGA-II to optimize task assignment. A Pareto-based analysis is then conducted to compare the performance of the conventional A* approach with that of the proposed A*&amp;amp;ndash;NSGA-II framework, highlighting the trade-offs among the selected operational objectives. The findings demonstrate that shorter travel distances or earlier task completion do not necessarily imply lower energy consumption, highlighting the need to select planning methods according to warehouse geometry, fleet workload, and operational priorities.</description>
	<pubDate>2026-11-18</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 191: Multi-Objective AGV Task Allocation and Path Planning: Evaluation of Deterministic and Metaheuristic Algorithms</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/191">doi: 10.3390/automation7060191</a></p>
	<p>Authors:
		Prema C. Thalaivar
		Madhusudhan Krishnagiri Narasimhamurthy
		Mamatha Mysore Nagarajan
		</p>
	<p>Automated guided vehicles (AGVs) enable material transport in automated manufacturing facilities, where fleet efficiency depends on coordinating task allocation and route generation while balancing completion time, energy consumption, and travel distance. This study develops a Python(version: 3.14.6)-based multi-objective evaluation framework for offline AGV path planning and task assignment in grid-based warehouse environments with known, static obstacles. Particle swarm optimization (PSO) and the gravitational search algorithm (GSA) are evaluated against deterministic A* search across warehouse scenarios with different layouts and obstacle configurations. Each planner is integrated into a common greedy assignment procedure that processes tasks according to priority and earliest-start ordering. For each task, candidate AGVs are evaluated using planned routes from their current positions to the pickup location and subsequently to the delivery location. Assignment decisions minimize a weighted cost combining estimated completion time, route energy consumption, and travel distance. Fleet-level performance is assessed using makespan, defined as the maximum completion time across AGVs, together with aggregate energy consumption and total travel distance. Energy estimates account for motion-related costs, including acceleration and turning, enabling comparisons beyond geometric path length alone. The benchmark results indicate that A* provides consistently competitive performance, particularly in energy consumption, whereas PSO achieves favorable makespan&amp;amp;ndash;distance trade-offs in selected scenarios. As a future extension toward online path planning, the A* algorithm, which supports real-time path generation, is integrated with NSGA-II to optimize task assignment. A Pareto-based analysis is then conducted to compare the performance of the conventional A* approach with that of the proposed A*&amp;amp;ndash;NSGA-II framework, highlighting the trade-offs among the selected operational objectives. The findings demonstrate that shorter travel distances or earlier task completion do not necessarily imply lower energy consumption, highlighting the need to select planning methods according to warehouse geometry, fleet workload, and operational priorities.</p>
	]]></content:encoded>

	<dc:title>Multi-Objective AGV Task Allocation and Path Planning: Evaluation of Deterministic and Metaheuristic Algorithms</dc:title>
			<dc:creator>Prema C. Thalaivar</dc:creator>
			<dc:creator>Madhusudhan Krishnagiri Narasimhamurthy</dc:creator>
			<dc:creator>Mamatha Mysore Nagarajan</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060191</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-11-18</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-11-18</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>191</prism:startingPage>
		<prism:doi>10.3390/automation7060191</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/191</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/190">

	<title>Automation, Vol. 7, Pages 190: Research and Development of Vibration Diagnostics Methods for Electric Motors Using Machine Learning</title>
	<link>https://www.mdpi.com/2673-4052/7/6/190</link>
	<description>Electric motors are the principal power units of industry, and their failures cause cascading stoppages of process equipment. The vibration signal carries the most complete information about the condition of rotating assemblies, yet machine-learning diagnostics is constrained by the shortage of labelled fault data. This work proposes and verifies a methodology for generating a physically grounded synthetic dataset of vibration signals: analytical models of imbalance, misalignment, rolling-bearing defects (BPFO, BPFI, BSF, FTF) and mechanical looseness produced 8000 one-second fragments of vibration acceleration covering eight classes of technical state, with variation of the rotational frequency (&amp;amp;plusmn;5%), signal-to-noise ratio (5&amp;amp;ndash;30 dB) and defect amplitude. The synthetic signals were verified against reference Case Western Reserve University (CWRU) records: the outer- and inner-ring defect frequencies agree within 0.22%, and the envelope spectra are class-specific for these defects. Six classifiers were compared in three stages: Random Forest, XGBoost, a 1D-CNN (WDCNN), two 2D-CNNs on STFT and CWT representations, and a PatchTST transformer. Under the full set of variations, models reached a Macro F1 of 99.6&amp;amp;ndash;99.8%. On the real CWRU records, separate models were trained on the dataset based on real data from other operating loads. On the unseen loads, the Random Forest reached a Macro F1 of 95.5% and the 1D-CNN 93.2%, against 21&amp;amp;ndash;42% for the same models trained on synthetic data alone, and with one real fragment per class, the difference between the two training datasets was already 17&amp;amp;ndash;32 p.p. High accuracy on synthetic data therefore did not carry over to zero-shot diagnosis on the real bench.</description>
	<pubDate>2026-11-16</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 190: Research and Development of Vibration Diagnostics Methods for Electric Motors Using Machine Learning</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/190">doi: 10.3390/automation7060190</a></p>
	<p>Authors:
		Yuriy Kozhubaev
		Dmitry Kazanin
		Roman Dorovskikh
		Pavel Dolgiy
		Georgy Molodtsov
		</p>
	<p>Electric motors are the principal power units of industry, and their failures cause cascading stoppages of process equipment. The vibration signal carries the most complete information about the condition of rotating assemblies, yet machine-learning diagnostics is constrained by the shortage of labelled fault data. This work proposes and verifies a methodology for generating a physically grounded synthetic dataset of vibration signals: analytical models of imbalance, misalignment, rolling-bearing defects (BPFO, BPFI, BSF, FTF) and mechanical looseness produced 8000 one-second fragments of vibration acceleration covering eight classes of technical state, with variation of the rotational frequency (&amp;amp;plusmn;5%), signal-to-noise ratio (5&amp;amp;ndash;30 dB) and defect amplitude. The synthetic signals were verified against reference Case Western Reserve University (CWRU) records: the outer- and inner-ring defect frequencies agree within 0.22%, and the envelope spectra are class-specific for these defects. Six classifiers were compared in three stages: Random Forest, XGBoost, a 1D-CNN (WDCNN), two 2D-CNNs on STFT and CWT representations, and a PatchTST transformer. Under the full set of variations, models reached a Macro F1 of 99.6&amp;amp;ndash;99.8%. On the real CWRU records, separate models were trained on the dataset based on real data from other operating loads. On the unseen loads, the Random Forest reached a Macro F1 of 95.5% and the 1D-CNN 93.2%, against 21&amp;amp;ndash;42% for the same models trained on synthetic data alone, and with one real fragment per class, the difference between the two training datasets was already 17&amp;amp;ndash;32 p.p. High accuracy on synthetic data therefore did not carry over to zero-shot diagnosis on the real bench.</p>
	]]></content:encoded>

	<dc:title>Research and Development of Vibration Diagnostics Methods for Electric Motors Using Machine Learning</dc:title>
			<dc:creator>Yuriy Kozhubaev</dc:creator>
			<dc:creator>Dmitry Kazanin</dc:creator>
			<dc:creator>Roman Dorovskikh</dc:creator>
			<dc:creator>Pavel Dolgiy</dc:creator>
			<dc:creator>Georgy Molodtsov</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060190</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-11-16</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-11-16</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>190</prism:startingPage>
		<prism:doi>10.3390/automation7060190</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/190</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/189">

	<title>Automation, Vol. 7, Pages 189: Two-Stage Supervisory MPC for Cooperative UAV Coverage with Fault-Triggered Mission Reconfiguration</title>
	<link>https://www.mdpi.com/2673-4052/7/6/189</link>
	<description>A two-stage supervisory model predictive control (MPC) framework is developed for cooperative unmanned aerial vehicle (UAV) coverage during mission transit and reconfiguration after single-agent loss. A synchronized predictive supervisory layer coordinates sector responsibility and constrained motion planning through a ranked assignment&amp;amp;ndash;feasibility architecture: Stage 1 orders candidate sector assignments, and Stage 2 selects an assignment-consistent finite-horizon trajectory. This explicitly links combinatorial task allocation to predictive sensing-service constraints while separating mission coordination from low-level vehicle stabilization. Existence, exact-model interval-wise constraint satisfaction, and conditional post-fault solvability are characterized for the finite-dimensional analytical formulation. A fully specified, reproducible finite-control-library realization is assessed through matched-weight nominal ablations, continuous-time separation checks on executed first-order trajectories, and post-fault task-coverage histories. An additional controlled study tests assignment-conditioned trajectory candidates against frozen-task and universal-library references for five deterministic fault configurations, with matched pre-fault motion within each comparison. The experiments demonstrate a reproducible post-fault redistribution mechanism: candidate responsibilities are coupled to predictive service-feasibility tests, and the resulting motion choices are evaluated against coverage, overlap, terminal progress, and computation. Matched ablations and five paired fault configurations quantify both the sensing effects and the competing objectives of the supervisory policy. A supplementary two-phase completion experiment then switches from the monitoring task to a terminal gathering mode and shows representative nominal and post-fault arrivals to the target region under the same first-order safety model.</description>
	<pubDate>2026-11-15</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 189: Two-Stage Supervisory MPC for Cooperative UAV Coverage with Fault-Triggered Mission Reconfiguration</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/189">doi: 10.3390/automation7060189</a></p>
	<p>Authors:
		Robert Vrabel
		</p>
	<p>A two-stage supervisory model predictive control (MPC) framework is developed for cooperative unmanned aerial vehicle (UAV) coverage during mission transit and reconfiguration after single-agent loss. A synchronized predictive supervisory layer coordinates sector responsibility and constrained motion planning through a ranked assignment&amp;amp;ndash;feasibility architecture: Stage 1 orders candidate sector assignments, and Stage 2 selects an assignment-consistent finite-horizon trajectory. This explicitly links combinatorial task allocation to predictive sensing-service constraints while separating mission coordination from low-level vehicle stabilization. Existence, exact-model interval-wise constraint satisfaction, and conditional post-fault solvability are characterized for the finite-dimensional analytical formulation. A fully specified, reproducible finite-control-library realization is assessed through matched-weight nominal ablations, continuous-time separation checks on executed first-order trajectories, and post-fault task-coverage histories. An additional controlled study tests assignment-conditioned trajectory candidates against frozen-task and universal-library references for five deterministic fault configurations, with matched pre-fault motion within each comparison. The experiments demonstrate a reproducible post-fault redistribution mechanism: candidate responsibilities are coupled to predictive service-feasibility tests, and the resulting motion choices are evaluated against coverage, overlap, terminal progress, and computation. Matched ablations and five paired fault configurations quantify both the sensing effects and the competing objectives of the supervisory policy. A supplementary two-phase completion experiment then switches from the monitoring task to a terminal gathering mode and shows representative nominal and post-fault arrivals to the target region under the same first-order safety model.</p>
	]]></content:encoded>

	<dc:title>Two-Stage Supervisory MPC for Cooperative UAV Coverage with Fault-Triggered Mission Reconfiguration</dc:title>
			<dc:creator>Robert Vrabel</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060189</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-11-15</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-11-15</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>189</prism:startingPage>
		<prism:doi>10.3390/automation7060189</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/189</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/188">

	<title>Automation, Vol. 7, Pages 188: Calibration for a Redundant Serial Robot Based on an Elasto-Geometrical Model and an Adaptive Identification Algorithm</title>
	<link>https://www.mdpi.com/2673-4052/7/6/188</link>
	<description>This paper proposes a novel elasto-geometrical coupling calibration and an adaptive parameter identification method for flexible redundant manipulators. For the modeling, an elasto-geometrical coupling calibration model containing kinematic parameter errors and non-kinematic errors is constructed. Secondly, an adaptive identification rule is formulated to suppress the elasto-geometrical modeling errors according to the residual linearization errors in each iteration, thus obtaining accurate identification results even in the case of highly elastic joints. Thirdly, a dual-target measuring strategy based on multi-source sensing is designed to obtain high-precision hand-eye data; thus, lightweight sensors and a vision system suitable for flexible light-load manipulators can be practically used for the calibration. As revealed by simulation tests, the newly developed identification algorithm can progressively compensate for considerable linearization modeling discrepancies for a redundant robotic arm with 12 highly elastic joints. All error parameters are estimated with an accuracy above 99% through iterative updates. Corresponding experiments are carried out on an eight-joint cable-driven elastic redundant manipulator, and the results indicate the new calibration method reaches the fastest algorithm convergence and residuals at the same level when compared with two traditional ones.</description>
	<pubDate>2026-11-13</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 188: Calibration for a Redundant Serial Robot Based on an Elasto-Geometrical Model and an Adaptive Identification Algorithm</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/188">doi: 10.3390/automation7060188</a></p>
	<p>Authors:
		Bin Wang
		Zhouxiang Jiang
		Shuhui Xie
		</p>
	<p>This paper proposes a novel elasto-geometrical coupling calibration and an adaptive parameter identification method for flexible redundant manipulators. For the modeling, an elasto-geometrical coupling calibration model containing kinematic parameter errors and non-kinematic errors is constructed. Secondly, an adaptive identification rule is formulated to suppress the elasto-geometrical modeling errors according to the residual linearization errors in each iteration, thus obtaining accurate identification results even in the case of highly elastic joints. Thirdly, a dual-target measuring strategy based on multi-source sensing is designed to obtain high-precision hand-eye data; thus, lightweight sensors and a vision system suitable for flexible light-load manipulators can be practically used for the calibration. As revealed by simulation tests, the newly developed identification algorithm can progressively compensate for considerable linearization modeling discrepancies for a redundant robotic arm with 12 highly elastic joints. All error parameters are estimated with an accuracy above 99% through iterative updates. Corresponding experiments are carried out on an eight-joint cable-driven elastic redundant manipulator, and the results indicate the new calibration method reaches the fastest algorithm convergence and residuals at the same level when compared with two traditional ones.</p>
	]]></content:encoded>

	<dc:title>Calibration for a Redundant Serial Robot Based on an Elasto-Geometrical Model and an Adaptive Identification Algorithm</dc:title>
			<dc:creator>Bin Wang</dc:creator>
			<dc:creator>Zhouxiang Jiang</dc:creator>
			<dc:creator>Shuhui Xie</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060188</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-11-13</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-11-13</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>188</prism:startingPage>
		<prism:doi>10.3390/automation7060188</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/188</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/187">

	<title>Automation, Vol. 7, Pages 187: Performance- and Mechanical-Work-Aware Control Reconfiguration of a Mecanum-Wheeled Mobile Robot After Complete Wheel Actuator Failure</title>
	<link>https://www.mdpi.com/2673-4052/7/6/187</link>
	<description>Wheel-actuator failure can compromise the trajectory-tracking capability of omnidirectional mobile robots, potentially interrupting material transport in automated warehouses. This study investigates control allocation reconfiguration to maintain trajectory tracking of a four-Mecanum-wheeled mobile robot after a single wheel actuator completely loses its torque-generation capability. Assuming that the failed actuator has already been detected and isolated, the proposed approach retains the nominal trajectory-tracking controller and redistributes the required generalized forces and yaw moment among the remaining actuators. The feasibility of post-fault tracking is examined through the rank of the available allocation matrix and the actuator torque limits. Five operating configurations are compared in simulation, including healthy operation, unreconfigured operation after failure, three-actuator reconfiguration, and two reduced two-actuator configurations. The achieved results demonstrate that, under the investigated simulation conditions, actuator-aware control allocation can preserve both position and orientation tracking after a single actuator failure without retuning the nominal controller, supporting continued robot motion during a post-fault operating interval.</description>
	<pubDate>2026-11-12</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 187: Performance- and Mechanical-Work-Aware Control Reconfiguration of a Mecanum-Wheeled Mobile Robot After Complete Wheel Actuator Failure</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/187">doi: 10.3390/automation7060187</a></p>
	<p>Authors:
		Dinh Anh Nguyen
		Viet Phuong Pham
		Manh Linh Nguyen
		</p>
	<p>Wheel-actuator failure can compromise the trajectory-tracking capability of omnidirectional mobile robots, potentially interrupting material transport in automated warehouses. This study investigates control allocation reconfiguration to maintain trajectory tracking of a four-Mecanum-wheeled mobile robot after a single wheel actuator completely loses its torque-generation capability. Assuming that the failed actuator has already been detected and isolated, the proposed approach retains the nominal trajectory-tracking controller and redistributes the required generalized forces and yaw moment among the remaining actuators. The feasibility of post-fault tracking is examined through the rank of the available allocation matrix and the actuator torque limits. Five operating configurations are compared in simulation, including healthy operation, unreconfigured operation after failure, three-actuator reconfiguration, and two reduced two-actuator configurations. The achieved results demonstrate that, under the investigated simulation conditions, actuator-aware control allocation can preserve both position and orientation tracking after a single actuator failure without retuning the nominal controller, supporting continued robot motion during a post-fault operating interval.</p>
	]]></content:encoded>

	<dc:title>Performance- and Mechanical-Work-Aware Control Reconfiguration of a Mecanum-Wheeled Mobile Robot After Complete Wheel Actuator Failure</dc:title>
			<dc:creator>Dinh Anh Nguyen</dc:creator>
			<dc:creator>Viet Phuong Pham</dc:creator>
			<dc:creator>Manh Linh Nguyen</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060187</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-11-12</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-11-12</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>187</prism:startingPage>
		<prism:doi>10.3390/automation7060187</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/187</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/186">

	<title>Automation, Vol. 7, Pages 186: An Engineering Neuro-Fuzzy Decision Support System for Optimizing an Industrial Sulfur Production Process Under Uncertainty</title>
	<link>https://www.mdpi.com/2673-4052/7/6/186</link>
	<description>Sulfur recovery units (SRUs) at oil refineries are complex, multi-parameter systems whose Cold Bed Absorption (CBA) reactors operate under incomplete and uncertain process information, which limits conventional control. This study develops an engineering neuro-fuzzy decision support system (DSS) for optimizing sulfur production at the Atyrau oil refinery (Kazakhstan) under uncertainty. The system integrates a mathematical description of the process, whose dynamic part is stated as a general formulation, an expert-derived fuzzy rule base, a first-order Sugeno neuro-fuzzy estimator of sulfur yield and quality, and a multi-criteria fuzzy optimization block based on the Bellman&amp;amp;ndash;Zadeh approach, together with an explicit operator-facing decision algorithm and a software interface. The estimators were identified and validated on 200 operating log records of the unit. These records show that the effect of reactor temperature on the sulfur quality index reverses sign with reactor pressure&amp;amp;mdash;the correlation is &amp;amp;minus;0.57 below 0.79 kgf/cm2 and +0.75 above it&amp;amp;mdash;so that a model additive in the two variables cannot represent the relationship: linear regression on temperature and pressure attains a cross-validated R2 of only 0.06 for the quality index, whereas the neuro-fuzzy estimator attains 0.83 (RMSE 0.72 points, MAPE 0.74%). A regression with an explicit interaction term on the same inputs attains 0.79; in a paired comparison the neuro-fuzzy estimator had the lower error in every repetition, but the difference is not statistically significant (p = 0.07), and the estimator is retained for its rule-level interpretability and its integration with the decision block. For sulfur yield, where the relationship is close to linear once pressure and throughput are accounted for, the corresponding figures are 0.71 and 0.85 (RMSE 0.33 t/day, MAPE 1.35%). The same ordering is reproduced on an independent hold-out test set, under nested and cluster-blocked cross-validation and after removal of the records lying at the unit&amp;amp;rsquo;s capacity and regulatory bounds, and the expert rule base was revised into an asymmetric form accordingly. A worked application of the optimization block to three operating situations, with the production plan, quality and capacity constraints enforced, raised the model-predicted degree of joint satisfaction of the yield, quality and energy criteria from 0.08&amp;amp;ndash;0.25 to 0.49&amp;amp;ndash;0.57; in one of the three, the recommended action is to reduce process severity and lower a reheat duty surrogate by about a third rather than to increase output. The resulting DSS, at present a standalone advisory software prototype, supports the operator in analyzing the current process state, evaluating alternative operating modes and forming recommendations, while the final decision remains with the operator.</description>
	<pubDate>2026-11-11</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 186: An Engineering Neuro-Fuzzy Decision Support System for Optimizing an Industrial Sulfur Production Process Under Uncertainty</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/186">doi: 10.3390/automation7060186</a></p>
	<p>Authors:
		Batyr Orazbayev
		Madyar Kabibullin
		Narkez Boranbayeva
		Balbupe Utenova
		Kulman Orazbayeva
		Almagul Yessengaliyeva
		Araigul Tazhgaliyeva
		</p>
	<p>Sulfur recovery units (SRUs) at oil refineries are complex, multi-parameter systems whose Cold Bed Absorption (CBA) reactors operate under incomplete and uncertain process information, which limits conventional control. This study develops an engineering neuro-fuzzy decision support system (DSS) for optimizing sulfur production at the Atyrau oil refinery (Kazakhstan) under uncertainty. The system integrates a mathematical description of the process, whose dynamic part is stated as a general formulation, an expert-derived fuzzy rule base, a first-order Sugeno neuro-fuzzy estimator of sulfur yield and quality, and a multi-criteria fuzzy optimization block based on the Bellman&amp;amp;ndash;Zadeh approach, together with an explicit operator-facing decision algorithm and a software interface. The estimators were identified and validated on 200 operating log records of the unit. These records show that the effect of reactor temperature on the sulfur quality index reverses sign with reactor pressure&amp;amp;mdash;the correlation is &amp;amp;minus;0.57 below 0.79 kgf/cm2 and +0.75 above it&amp;amp;mdash;so that a model additive in the two variables cannot represent the relationship: linear regression on temperature and pressure attains a cross-validated R2 of only 0.06 for the quality index, whereas the neuro-fuzzy estimator attains 0.83 (RMSE 0.72 points, MAPE 0.74%). A regression with an explicit interaction term on the same inputs attains 0.79; in a paired comparison the neuro-fuzzy estimator had the lower error in every repetition, but the difference is not statistically significant (p = 0.07), and the estimator is retained for its rule-level interpretability and its integration with the decision block. For sulfur yield, where the relationship is close to linear once pressure and throughput are accounted for, the corresponding figures are 0.71 and 0.85 (RMSE 0.33 t/day, MAPE 1.35%). The same ordering is reproduced on an independent hold-out test set, under nested and cluster-blocked cross-validation and after removal of the records lying at the unit&amp;amp;rsquo;s capacity and regulatory bounds, and the expert rule base was revised into an asymmetric form accordingly. A worked application of the optimization block to three operating situations, with the production plan, quality and capacity constraints enforced, raised the model-predicted degree of joint satisfaction of the yield, quality and energy criteria from 0.08&amp;amp;ndash;0.25 to 0.49&amp;amp;ndash;0.57; in one of the three, the recommended action is to reduce process severity and lower a reheat duty surrogate by about a third rather than to increase output. The resulting DSS, at present a standalone advisory software prototype, supports the operator in analyzing the current process state, evaluating alternative operating modes and forming recommendations, while the final decision remains with the operator.</p>
	]]></content:encoded>

	<dc:title>An Engineering Neuro-Fuzzy Decision Support System for Optimizing an Industrial Sulfur Production Process Under Uncertainty</dc:title>
			<dc:creator>Batyr Orazbayev</dc:creator>
			<dc:creator>Madyar Kabibullin</dc:creator>
			<dc:creator>Narkez Boranbayeva</dc:creator>
			<dc:creator>Balbupe Utenova</dc:creator>
			<dc:creator>Kulman Orazbayeva</dc:creator>
			<dc:creator>Almagul Yessengaliyeva</dc:creator>
			<dc:creator>Araigul Tazhgaliyeva</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060186</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-11-11</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-11-11</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>186</prism:startingPage>
		<prism:doi>10.3390/automation7060186</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/186</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/185">

	<title>Automation, Vol. 7, Pages 185: Hybrid Quantum-Classical Machine Learning for Air Quality Prediction in Indian Cities</title>
	<link>https://www.mdpi.com/2673-4052/7/6/185</link>
	<description>Short-term air-quality forecasting is important for environmental monitoring and public-health planning, particularly in Indian cities affected by persistent air pollution. This study presents a next-day air-quality prediction framework for seven pollutants (PM2.5, PM10, NO2, SO2, CO, NOx, and O3) across 11 Indian cities over the period 2019&amp;amp;ndash;2023, using daily observations from the Central Pollution Control Board (CPCB) and meteorological data from NASA POWER. Next-day pollutant concentrations are predicted using historical pollutant measurements, lagged and rolling features, calendar variables, and meteorological covariates available at the feature date. The study documents Variational Quantum Circuit (VQC) configurations implemented using PennyLane and Qiskit, alongside a candidate quantum-feature-augmentation approach for XGBoost. Quantitative evaluation is restricted to the XGBoost model, using mean absolute error (MAE), root mean squared error (RMSE), and the coefficient of determination R2 over the common chronological final-test target window spanning 28 June 2022 to 31 March 2023. For the evaluated XGBoost predictions, the reported pollutant-specific R2 values range from 0.613 to 0.864. The quantum-circuit configurations and quantum-feature-augmentation approach are presented as candidate methods whose comparative predictive performance remains to be established through evaluation using aligned final-test predictions and observed targets.</description>
	<pubDate>2026-11-11</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 185: Hybrid Quantum-Classical Machine Learning for Air Quality Prediction in Indian Cities</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/185">doi: 10.3390/automation7060185</a></p>
	<p>Authors:
		S. Baghavathi Priya
		Krithikha Sanju Saravanan
		Dinesh Sirigineedi
		Hiroshini Kannan
		Marabattini Sumana
		</p>
	<p>Short-term air-quality forecasting is important for environmental monitoring and public-health planning, particularly in Indian cities affected by persistent air pollution. This study presents a next-day air-quality prediction framework for seven pollutants (PM2.5, PM10, NO2, SO2, CO, NOx, and O3) across 11 Indian cities over the period 2019&amp;amp;ndash;2023, using daily observations from the Central Pollution Control Board (CPCB) and meteorological data from NASA POWER. Next-day pollutant concentrations are predicted using historical pollutant measurements, lagged and rolling features, calendar variables, and meteorological covariates available at the feature date. The study documents Variational Quantum Circuit (VQC) configurations implemented using PennyLane and Qiskit, alongside a candidate quantum-feature-augmentation approach for XGBoost. Quantitative evaluation is restricted to the XGBoost model, using mean absolute error (MAE), root mean squared error (RMSE), and the coefficient of determination R2 over the common chronological final-test target window spanning 28 June 2022 to 31 March 2023. For the evaluated XGBoost predictions, the reported pollutant-specific R2 values range from 0.613 to 0.864. The quantum-circuit configurations and quantum-feature-augmentation approach are presented as candidate methods whose comparative predictive performance remains to be established through evaluation using aligned final-test predictions and observed targets.</p>
	]]></content:encoded>

	<dc:title>Hybrid Quantum-Classical Machine Learning for Air Quality Prediction in Indian Cities</dc:title>
			<dc:creator>S. Baghavathi Priya</dc:creator>
			<dc:creator>Krithikha Sanju Saravanan</dc:creator>
			<dc:creator>Dinesh Sirigineedi</dc:creator>
			<dc:creator>Hiroshini Kannan</dc:creator>
			<dc:creator>Marabattini Sumana</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060185</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-11-11</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-11-11</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>185</prism:startingPage>
		<prism:doi>10.3390/automation7060185</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/185</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/184">

	<title>Automation, Vol. 7, Pages 184: An Adaptive Reconstruction and Branch-Interaction Network for False Data Injection Attack Detection in Smart Grids</title>
	<link>https://www.mdpi.com/2673-4052/7/6/184</link>
	<description>False data injection attacks (FDIAs) pose a serious threat to smart grids because they can alter state estimates without triggering conventional bad-data detection. Although learning-based detectors reduce the need for an exact system model, many of them focus on either temporal changes or relations among measurement variables. To address this problem, this paper presents an Adaptive Reconstruction and Branch-Interaction Network (ARBI-Net). A bidirectional long short-term memory network extracts temporal features from measurement windows, while an inverted Transformer learns relations among electrical variables. Between the two branches, a variable-token reconstruction module converts the temporal representations into tokens linked to individual measurement channels. An adaptive fusion module then combines the temporal and cross-variable features for classification. Experiments are conducted on the IEEE 6-bus, 14-bus, and 118-bus systems. The ablation study shows that the complete model achieves an F1-score of 97.18%, compared with 96.98% for BiLSTM alone and 94.39% for iTransformer alone. On the IEEE 14-bus and 118-bus systems, ARBI-Net achieves F1-scores of 92.60% and 98.50%, exceeding the best learning-based baselines by 1.41 and 10.63 percentage points, respectively. It also provides a better balance between missed detections and false alarms. Further tests show that its detection performance remains more stable under changes in attack strength and measurement noise.</description>
	<pubDate>2026-11-10</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 184: An Adaptive Reconstruction and Branch-Interaction Network for False Data Injection Attack Detection in Smart Grids</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/184">doi: 10.3390/automation7060184</a></p>
	<p>Authors:
		Shuxiang Wang
		Xinyu Wang
		</p>
	<p>False data injection attacks (FDIAs) pose a serious threat to smart grids because they can alter state estimates without triggering conventional bad-data detection. Although learning-based detectors reduce the need for an exact system model, many of them focus on either temporal changes or relations among measurement variables. To address this problem, this paper presents an Adaptive Reconstruction and Branch-Interaction Network (ARBI-Net). A bidirectional long short-term memory network extracts temporal features from measurement windows, while an inverted Transformer learns relations among electrical variables. Between the two branches, a variable-token reconstruction module converts the temporal representations into tokens linked to individual measurement channels. An adaptive fusion module then combines the temporal and cross-variable features for classification. Experiments are conducted on the IEEE 6-bus, 14-bus, and 118-bus systems. The ablation study shows that the complete model achieves an F1-score of 97.18%, compared with 96.98% for BiLSTM alone and 94.39% for iTransformer alone. On the IEEE 14-bus and 118-bus systems, ARBI-Net achieves F1-scores of 92.60% and 98.50%, exceeding the best learning-based baselines by 1.41 and 10.63 percentage points, respectively. It also provides a better balance between missed detections and false alarms. Further tests show that its detection performance remains more stable under changes in attack strength and measurement noise.</p>
	]]></content:encoded>

	<dc:title>An Adaptive Reconstruction and Branch-Interaction Network for False Data Injection Attack Detection in Smart Grids</dc:title>
			<dc:creator>Shuxiang Wang</dc:creator>
			<dc:creator>Xinyu Wang</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060184</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-11-10</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-11-10</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>184</prism:startingPage>
		<prism:doi>10.3390/automation7060184</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/184</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/183">

	<title>Automation, Vol. 7, Pages 183: A Scalable Modular Hardware Architecture for Automated Testing of Automotive Electronic Control Units</title>
	<link>https://www.mdpi.com/2673-4052/7/6/183</link>
	<description>The increasing complexity of automotive electronic systems requires configurable laboratory platforms for Electronic Control Units (ECUs). The conventional test approaches, such as hardware-in-the-loop systems and customized test benches, may require substantial engineering effort when they are adapted to different ECU configurations. This research paper presents the implementation and initial functional laboratory validation of a modular hardware prototype for selected ECU I/O testing tasks. The system utilizes a backplane-based platform that combines application-specific hardware modules for communication, control, input stimulation, switching, and load interfacing. Coordinated operation of the implemented modules is provided by a central controller and a Classical CAN-based request&amp;amp;ndash;response protocol. The design allows compatible input and output modules to be replaced or added according to the required I/O configuration, subject to the mechanical, electrical, communication, power, and thermal limitations of the prototype. This research paper also outlines a test execution structure that includes initialization, input stimulation, load connection, module-status reporting, and predefined sequence execution. Initial functional validation was performed in a laboratory environment using a representative Body Control Module (BCM) test scenario. This research does not establish universal ECU compatibility, quantitative scalability, metrological measurement performance, industrial readiness, or equivalence to commercial HIL and modular instrumentation systems. The contribution is therefore limited to the implementation and initial functional validation of the described application-specific prototype.</description>
	<pubDate>2026-11-10</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 183: A Scalable Modular Hardware Architecture for Automated Testing of Automotive Electronic Control Units</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/183">doi: 10.3390/automation7060183</a></p>
	<p>Authors:
		Dumitru-Daniel Bonciog
		Adriana Berdich
		</p>
	<p>The increasing complexity of automotive electronic systems requires configurable laboratory platforms for Electronic Control Units (ECUs). The conventional test approaches, such as hardware-in-the-loop systems and customized test benches, may require substantial engineering effort when they are adapted to different ECU configurations. This research paper presents the implementation and initial functional laboratory validation of a modular hardware prototype for selected ECU I/O testing tasks. The system utilizes a backplane-based platform that combines application-specific hardware modules for communication, control, input stimulation, switching, and load interfacing. Coordinated operation of the implemented modules is provided by a central controller and a Classical CAN-based request&amp;amp;ndash;response protocol. The design allows compatible input and output modules to be replaced or added according to the required I/O configuration, subject to the mechanical, electrical, communication, power, and thermal limitations of the prototype. This research paper also outlines a test execution structure that includes initialization, input stimulation, load connection, module-status reporting, and predefined sequence execution. Initial functional validation was performed in a laboratory environment using a representative Body Control Module (BCM) test scenario. This research does not establish universal ECU compatibility, quantitative scalability, metrological measurement performance, industrial readiness, or equivalence to commercial HIL and modular instrumentation systems. The contribution is therefore limited to the implementation and initial functional validation of the described application-specific prototype.</p>
	]]></content:encoded>

	<dc:title>A Scalable Modular Hardware Architecture for Automated Testing of Automotive Electronic Control Units</dc:title>
			<dc:creator>Dumitru-Daniel Bonciog</dc:creator>
			<dc:creator>Adriana Berdich</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060183</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-11-10</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-11-10</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>183</prism:startingPage>
		<prism:doi>10.3390/automation7060183</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/183</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/182">

	<title>Automation, Vol. 7, Pages 182: A Real-Time Unity Framework for Integrated High- and Low-Level Drone Control: Standoff Object-Tracking Strategy as a Use Case</title>
	<link>https://www.mdpi.com/2673-4052/7/6/182</link>
	<description>Developing multirotor control systems can require separate tools for vehicle dynamics, low-level control, high-level autonomy, visualization, and data collection. This paper presents a modular framework implemented natively in Unity to support experiments examining interactions across these layers within one standalone environment. Implemented in C#, the framework combines an individual-rotor multirotor model, a quaternion-based cascaded low-level controller, a parametric trajectory generator, runtime parameter controls, and telemetry/data export. A position-based visual servoing extension implements standoff object tracking. The framework is evaluated through step-response and trajectory-following experiments, comparisons of dead-zone and fixed-radius tracking, and execution profiling. In the tested tracking scenarios, the fixed-radius approach maintained spatial constraints more strictly during smooth target motion, whereas the dead-zone approach reduced unnecessary corrections and generally improved screen-space stability, particularly around abrupt direction changes. Screen-space evaluation used target positions supplied by Unity ground truth; image-based detection was not assessed. Real-time stress tests on the tested laptop showed that the standalone builds maintained an approximately 1:1 real-time factor across the tested 0.5&amp;amp;ndash;10 kHz range; their frame rates reached the 30 FPS usability threshold at approximately 9 kHz for IL2CPP and 8.5 kHz for Mono. The results demonstrate the feasibility of using Unity for integrated multilevel-control prototyping and experimentation. An interactive WebGL demonstration with runtime parameter adjustment and telemetry is publicly available online.</description>
	<pubDate>2026-11-08</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 182: A Real-Time Unity Framework for Integrated High- and Low-Level Drone Control: Standoff Object-Tracking Strategy as a Use Case</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/182">doi: 10.3390/automation7060182</a></p>
	<p>Authors:
		Abdelghani Zabel
		Ibrahim Farouk Bouguenna
		Ahmed Tahour
		</p>
	<p>Developing multirotor control systems can require separate tools for vehicle dynamics, low-level control, high-level autonomy, visualization, and data collection. This paper presents a modular framework implemented natively in Unity to support experiments examining interactions across these layers within one standalone environment. Implemented in C#, the framework combines an individual-rotor multirotor model, a quaternion-based cascaded low-level controller, a parametric trajectory generator, runtime parameter controls, and telemetry/data export. A position-based visual servoing extension implements standoff object tracking. The framework is evaluated through step-response and trajectory-following experiments, comparisons of dead-zone and fixed-radius tracking, and execution profiling. In the tested tracking scenarios, the fixed-radius approach maintained spatial constraints more strictly during smooth target motion, whereas the dead-zone approach reduced unnecessary corrections and generally improved screen-space stability, particularly around abrupt direction changes. Screen-space evaluation used target positions supplied by Unity ground truth; image-based detection was not assessed. Real-time stress tests on the tested laptop showed that the standalone builds maintained an approximately 1:1 real-time factor across the tested 0.5&amp;amp;ndash;10 kHz range; their frame rates reached the 30 FPS usability threshold at approximately 9 kHz for IL2CPP and 8.5 kHz for Mono. The results demonstrate the feasibility of using Unity for integrated multilevel-control prototyping and experimentation. An interactive WebGL demonstration with runtime parameter adjustment and telemetry is publicly available online.</p>
	]]></content:encoded>

	<dc:title>A Real-Time Unity Framework for Integrated High- and Low-Level Drone Control: Standoff Object-Tracking Strategy as a Use Case</dc:title>
			<dc:creator>Abdelghani Zabel</dc:creator>
			<dc:creator>Ibrahim Farouk Bouguenna</dc:creator>
			<dc:creator>Ahmed Tahour</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060182</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-11-08</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-11-08</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>182</prism:startingPage>
		<prism:doi>10.3390/automation7060182</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/182</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/181">

	<title>Automation, Vol. 7, Pages 181: Adaptive and Explainable Hybrid Statistical&amp;ndash;Machine Learning Framework for Intelligent Industrial Process Monitoring</title>
	<link>https://www.mdpi.com/2673-4052/7/6/181</link>
	<description>An industrial monitor that changes its evidence weights also changes the rule that must be calibrated and explained. We combine principal-component residual energy, latent-space multivariate exponentially weighted moving averages, and a boosted-tree corruption detector in a fixed-reference monitoring policy. Past evidence determines the weights, complete episodes calibrate the persistent alarm rule, and an additive audit reconstructs each decision. Eight independently fitted simulation experiments give mean capped delays of 158.3 samples for hybrid adaptation, 180.8 for equal-weight fusion, and 147.5 for tuned static fusion. Controlled substitutions preserve the channel count and weight path: adding learned evidence increases delay by 15.7 samples at a common limit, but reduces it by 15.3 after separate calibration. The benefit therefore depends on the interaction between evidence and calibration. The evaluation includes isolation-forest comparators and explanation tests across nine scenarios. Stable attributions do not consistently identify perturbed variables. Continuous DAMADICS replay produces seven clean signals during 16 documented interventions and 27.2% background alarm activity. Actuator-2 distribution shifts persist within coarse controller-output strata, but their operational cause remains unknown. The framework supports auditable decisions and finite-episode calibration under exchangeability; it does not establish physical diagnosis or continuous-stream alarm control.</description>
	<pubDate>2026-11-06</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 181: Adaptive and Explainable Hybrid Statistical&amp;ndash;Machine Learning Framework for Intelligent Industrial Process Monitoring</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/181">doi: 10.3390/automation7060181</a></p>
	<p>Authors:
		Manuel J. Campuzano
		Rick Acosta-Vega
		Néstor Caicedo-Solano
		</p>
	<p>An industrial monitor that changes its evidence weights also changes the rule that must be calibrated and explained. We combine principal-component residual energy, latent-space multivariate exponentially weighted moving averages, and a boosted-tree corruption detector in a fixed-reference monitoring policy. Past evidence determines the weights, complete episodes calibrate the persistent alarm rule, and an additive audit reconstructs each decision. Eight independently fitted simulation experiments give mean capped delays of 158.3 samples for hybrid adaptation, 180.8 for equal-weight fusion, and 147.5 for tuned static fusion. Controlled substitutions preserve the channel count and weight path: adding learned evidence increases delay by 15.7 samples at a common limit, but reduces it by 15.3 after separate calibration. The benefit therefore depends on the interaction between evidence and calibration. The evaluation includes isolation-forest comparators and explanation tests across nine scenarios. Stable attributions do not consistently identify perturbed variables. Continuous DAMADICS replay produces seven clean signals during 16 documented interventions and 27.2% background alarm activity. Actuator-2 distribution shifts persist within coarse controller-output strata, but their operational cause remains unknown. The framework supports auditable decisions and finite-episode calibration under exchangeability; it does not establish physical diagnosis or continuous-stream alarm control.</p>
	]]></content:encoded>

	<dc:title>Adaptive and Explainable Hybrid Statistical&amp;amp;ndash;Machine Learning Framework for Intelligent Industrial Process Monitoring</dc:title>
			<dc:creator>Manuel J. Campuzano</dc:creator>
			<dc:creator>Rick Acosta-Vega</dc:creator>
			<dc:creator>Néstor Caicedo-Solano</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060181</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-11-06</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-11-06</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>181</prism:startingPage>
		<prism:doi>10.3390/automation7060181</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/181</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/180">

	<title>Automation, Vol. 7, Pages 180: A Spatiotemporal Transformer&amp;ndash;ConvLSTM Surrogate with Condition Embedding for Dynamic Three-Dimensional Temperature Field Prediction of XLPE HVDC Cables</title>
	<link>https://www.mdpi.com/2673-4052/7/6/180</link>
	<description>The conductor temperature of a high-voltage direct-current (HVDC) cable limits its current-carrying capacity and governs the aging of the cross-linked polyethylene (XLPE) insulation. Transient finite-element (FE) simulation provides the full temperature field but is too expensive for online rating and digital-twin applications. This paper presents an autoregressive deep-learning surrogate for the three-dimensional cable temperature field. Exploiting axisymmetry, the network predicts the next field on a two-dimensional radial&amp;amp;ndash;axial grid from the previous H frames and the operating condition, and the three-dimensional field is reconstructed by revolution. ConvLSTM layers capture spatiotemporal correlations, a temporal Transformer fuses the history states, and a continuous condition embedding lets one model cover the operating space. A scaled residual-increment formulation and a rollout curriculum stabilize the 600 min closed-loop prediction. On a COMSOL dataset of 49 operating conditions (ambient temperature &amp;amp;minus;20 to 40 &amp;amp;deg;C, load current 0 to 1200 A), the surrogate achieves RMSEs of 0.28, 0.27, and 0.49 &amp;amp;deg;C for training, interpolation, and extrapolation conditions, the latter being held-out corner combinations within the sweep ranges. It reduces the wall-clock time from about 110 s per FE simulation to 1.5 s per case (73&amp;amp;times; speed-up; 139&amp;amp;times; batched) and achieves lower mean RMSE than three baseline models.</description>
	<pubDate>2026-11-05</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 180: A Spatiotemporal Transformer&amp;ndash;ConvLSTM Surrogate with Condition Embedding for Dynamic Three-Dimensional Temperature Field Prediction of XLPE HVDC Cables</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/180">doi: 10.3390/automation7060180</a></p>
	<p>Authors:
		Kang-Jie Ruan
		Jia-Xun He
		Ya Zhang
		Hai-Yan Yang
		Ming-Liang Cheng
		Fan Xu
		Shi-Li Liu
		Wei Lu
		</p>
	<p>The conductor temperature of a high-voltage direct-current (HVDC) cable limits its current-carrying capacity and governs the aging of the cross-linked polyethylene (XLPE) insulation. Transient finite-element (FE) simulation provides the full temperature field but is too expensive for online rating and digital-twin applications. This paper presents an autoregressive deep-learning surrogate for the three-dimensional cable temperature field. Exploiting axisymmetry, the network predicts the next field on a two-dimensional radial&amp;amp;ndash;axial grid from the previous H frames and the operating condition, and the three-dimensional field is reconstructed by revolution. ConvLSTM layers capture spatiotemporal correlations, a temporal Transformer fuses the history states, and a continuous condition embedding lets one model cover the operating space. A scaled residual-increment formulation and a rollout curriculum stabilize the 600 min closed-loop prediction. On a COMSOL dataset of 49 operating conditions (ambient temperature &amp;amp;minus;20 to 40 &amp;amp;deg;C, load current 0 to 1200 A), the surrogate achieves RMSEs of 0.28, 0.27, and 0.49 &amp;amp;deg;C for training, interpolation, and extrapolation conditions, the latter being held-out corner combinations within the sweep ranges. It reduces the wall-clock time from about 110 s per FE simulation to 1.5 s per case (73&amp;amp;times; speed-up; 139&amp;amp;times; batched) and achieves lower mean RMSE than three baseline models.</p>
	]]></content:encoded>

	<dc:title>A Spatiotemporal Transformer&amp;amp;ndash;ConvLSTM Surrogate with Condition Embedding for Dynamic Three-Dimensional Temperature Field Prediction of XLPE HVDC Cables</dc:title>
			<dc:creator>Kang-Jie Ruan</dc:creator>
			<dc:creator>Jia-Xun He</dc:creator>
			<dc:creator>Ya Zhang</dc:creator>
			<dc:creator>Hai-Yan Yang</dc:creator>
			<dc:creator>Ming-Liang Cheng</dc:creator>
			<dc:creator>Fan Xu</dc:creator>
			<dc:creator>Shi-Li Liu</dc:creator>
			<dc:creator>Wei Lu</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060180</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-11-05</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-11-05</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>180</prism:startingPage>
		<prism:doi>10.3390/automation7060180</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/180</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/179">

	<title>Automation, Vol. 7, Pages 179: A Suitable-Area Selection Method for Road Vector Map Matching Based on a Hypergraph-Enhanced Graph Attention Network</title>
	<link>https://www.mdpi.com/2673-4052/7/6/179</link>
	<description>Suitable-area selection reduces candidate-search costs and mismatch risk in road vector map matching. Existing image-based assessments primarily characterize grayscale, texture, and edge saliency, whereas the suitability of road vector maps depends on road continuity, intersection organization, and topological uniqueness. This study proposes a high-confidence suitable-area selection framework based on a Hypergraph-Enhanced Graph Attention Network (HG-GAT). A road-network suitability dataset is constructed using offline matching results. Candidate regions are represented as road graphs, with semantic hyperedges encoding road chains, intersection stars, and key-node neighborhoods. Hypergraph propagation and graph attention jointly learn high-order structural relations and pairwise road connections, while dual-channel readout integrates regional organization and salient local structures. On the Wuhan test data, HG-GAT achieves an accuracy of 0.8074, an F1-score of 0.8309, and a Precision@10% of 0.8980, the highest values for these metrics among the compared methods. It retains high-confidence candidates with a recall of 0.9467 while reducing false selections relative to the image-based and ordinary graph-learning baselines. The results support the value of semantic high-order topology for distinguishing structurally effective road regions and prioritizing candidates for subsequent matching.</description>
	<pubDate>2026-11-03</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 179: A Suitable-Area Selection Method for Road Vector Map Matching Based on a Hypergraph-Enhanced Graph Attention Network</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/179">doi: 10.3390/automation7060179</a></p>
	<p>Authors:
		Hanyu Cai
		Yong Xian
		Shaopeng Li
		Decao Ma
		</p>
	<p>Suitable-area selection reduces candidate-search costs and mismatch risk in road vector map matching. Existing image-based assessments primarily characterize grayscale, texture, and edge saliency, whereas the suitability of road vector maps depends on road continuity, intersection organization, and topological uniqueness. This study proposes a high-confidence suitable-area selection framework based on a Hypergraph-Enhanced Graph Attention Network (HG-GAT). A road-network suitability dataset is constructed using offline matching results. Candidate regions are represented as road graphs, with semantic hyperedges encoding road chains, intersection stars, and key-node neighborhoods. Hypergraph propagation and graph attention jointly learn high-order structural relations and pairwise road connections, while dual-channel readout integrates regional organization and salient local structures. On the Wuhan test data, HG-GAT achieves an accuracy of 0.8074, an F1-score of 0.8309, and a Precision@10% of 0.8980, the highest values for these metrics among the compared methods. It retains high-confidence candidates with a recall of 0.9467 while reducing false selections relative to the image-based and ordinary graph-learning baselines. The results support the value of semantic high-order topology for distinguishing structurally effective road regions and prioritizing candidates for subsequent matching.</p>
	]]></content:encoded>

	<dc:title>A Suitable-Area Selection Method for Road Vector Map Matching Based on a Hypergraph-Enhanced Graph Attention Network</dc:title>
			<dc:creator>Hanyu Cai</dc:creator>
			<dc:creator>Yong Xian</dc:creator>
			<dc:creator>Shaopeng Li</dc:creator>
			<dc:creator>Decao Ma</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060179</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-11-03</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-11-03</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>179</prism:startingPage>
		<prism:doi>10.3390/automation7060179</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/179</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/178">

	<title>Automation, Vol. 7, Pages 178: Deduplicated Budget-Matched Benchmarking of Metaheuristic Feature Selection for PMSM Inverter Fault Diagnosis</title>
	<link>https://www.mdpi.com/2673-4052/7/6/178</link>
	<description>Metaheuristic feature-selection studies can overstate optimizer-specific effects when data duplicates, unequal evaluation budgets, weak baselines and representation choices are not controlled. This study therefore presents a deduplicated, budget-matched benchmark of six recent metaheuristics for permanent magnet synchronous motor (PMSM) inverter fault diagnosis rather than proposing a new optimizer or deployable diagnostic system. After removing 17 surplus exact-duplicate rows, the primary analysis uses 10,875 unique observations, class-wise blocked nested validation and an approximately 300-unique-evaluation budget, together with genetic algorithm (GA), particle swarm optimization (PSO), differential evolution (DE), random search, deterministic greedy selection and classifier-representation controls. The Runge&amp;amp;ndash;Kutta optimizer (RUN) achieved a mean macro-averaged F1 score (macro-F1) = 0.9473 with 4.20 features, while greedy forward selection achieved 0.9476 with 5.00 features; their paired fold difference was negligible (exact sign-flip p = 0.9375), and the six recent optimizers did not differ significantly at fold level (Friedman p = 0.2428). The evidence therefore supports sparse feature search but not metaheuristic superiority. Because each fault class originates from one approximately 10 Hz recording, the conclusions are restricted to blocked within-session benchmarking and do not establish cross-session, high-frequency or deployment reliability.</description>
	<pubDate>2026-11-02</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 178: Deduplicated Budget-Matched Benchmarking of Metaheuristic Feature Selection for PMSM Inverter Fault Diagnosis</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/178">doi: 10.3390/automation7060178</a></p>
	<p>Authors:
		Bonginkosi A. Thango
		Kumeshan Reddy
		</p>
	<p>Metaheuristic feature-selection studies can overstate optimizer-specific effects when data duplicates, unequal evaluation budgets, weak baselines and representation choices are not controlled. This study therefore presents a deduplicated, budget-matched benchmark of six recent metaheuristics for permanent magnet synchronous motor (PMSM) inverter fault diagnosis rather than proposing a new optimizer or deployable diagnostic system. After removing 17 surplus exact-duplicate rows, the primary analysis uses 10,875 unique observations, class-wise blocked nested validation and an approximately 300-unique-evaluation budget, together with genetic algorithm (GA), particle swarm optimization (PSO), differential evolution (DE), random search, deterministic greedy selection and classifier-representation controls. The Runge&amp;amp;ndash;Kutta optimizer (RUN) achieved a mean macro-averaged F1 score (macro-F1) = 0.9473 with 4.20 features, while greedy forward selection achieved 0.9476 with 5.00 features; their paired fold difference was negligible (exact sign-flip p = 0.9375), and the six recent optimizers did not differ significantly at fold level (Friedman p = 0.2428). The evidence therefore supports sparse feature search but not metaheuristic superiority. Because each fault class originates from one approximately 10 Hz recording, the conclusions are restricted to blocked within-session benchmarking and do not establish cross-session, high-frequency or deployment reliability.</p>
	]]></content:encoded>

	<dc:title>Deduplicated Budget-Matched Benchmarking of Metaheuristic Feature Selection for PMSM Inverter Fault Diagnosis</dc:title>
			<dc:creator>Bonginkosi A. Thango</dc:creator>
			<dc:creator>Kumeshan Reddy</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060178</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-11-02</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-11-02</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>178</prism:startingPage>
		<prism:doi>10.3390/automation7060178</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/178</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/177">

	<title>Automation, Vol. 7, Pages 177: An Integrated IoT-Based Smart Farming Framework for Tomato Cultivation Using Fuzzy Logic Irrigation Control and Deep Learning Classification</title>
	<link>https://www.mdpi.com/2673-4052/7/6/177</link>
	<description>Inefficient resource utilisation and changing environmental conditions pose significant challenges to tomato cultivation. This paper presents the design, implementation and experimental validation of a prototype smart farming system that combines Internet of Things (IoT) sensing, fuzzy-logic irrigation control and image-based classification of tomato status. An ESP32 microcontroller acquires soil-moisture, temperature, humidity and rainfall readings and publishes them through MQTT and HiveMQ Cloud to a PC, where a Mamdani fuzzy inference system implemented in MATLAB/Simulink R2025b computes the irrigation demand. The prototype was operated for four months with one potted plant. Separately, five classifiers were compared offline for tomato maturity classification. The five classifiers include ResNet18-based Convolutional Neural Networks (CNN), YOLOv8s cls, Random Forest, Support Vector Machine and K Nearest Neighbours. The ResNet18 achieved the highest test accuracy (95.00%; F1-score 94.97%), followed by YOLOv8s-cls (93.89%), whereas the classical models reached 85.6&amp;amp;ndash;88.1% on raw pixels and up to 90.8% on hand-crafted colour and texture features.</description>
	<pubDate>2026-11-01</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 177: An Integrated IoT-Based Smart Farming Framework for Tomato Cultivation Using Fuzzy Logic Irrigation Control and Deep Learning Classification</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/177">doi: 10.3390/automation7060177</a></p>
	<p>Authors:
		Anwar Sahbel
		Mohamed Reda
		Ehab H. E. Bayoumi
		Hoda Abuzied
		</p>
	<p>Inefficient resource utilisation and changing environmental conditions pose significant challenges to tomato cultivation. This paper presents the design, implementation and experimental validation of a prototype smart farming system that combines Internet of Things (IoT) sensing, fuzzy-logic irrigation control and image-based classification of tomato status. An ESP32 microcontroller acquires soil-moisture, temperature, humidity and rainfall readings and publishes them through MQTT and HiveMQ Cloud to a PC, where a Mamdani fuzzy inference system implemented in MATLAB/Simulink R2025b computes the irrigation demand. The prototype was operated for four months with one potted plant. Separately, five classifiers were compared offline for tomato maturity classification. The five classifiers include ResNet18-based Convolutional Neural Networks (CNN), YOLOv8s cls, Random Forest, Support Vector Machine and K Nearest Neighbours. The ResNet18 achieved the highest test accuracy (95.00%; F1-score 94.97%), followed by YOLOv8s-cls (93.89%), whereas the classical models reached 85.6&amp;amp;ndash;88.1% on raw pixels and up to 90.8% on hand-crafted colour and texture features.</p>
	]]></content:encoded>

	<dc:title>An Integrated IoT-Based Smart Farming Framework for Tomato Cultivation Using Fuzzy Logic Irrigation Control and Deep Learning Classification</dc:title>
			<dc:creator>Anwar Sahbel</dc:creator>
			<dc:creator>Mohamed Reda</dc:creator>
			<dc:creator>Ehab H. E. Bayoumi</dc:creator>
			<dc:creator>Hoda Abuzied</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060177</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-11-01</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-11-01</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>177</prism:startingPage>
		<prism:doi>10.3390/automation7060177</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/177</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/176">

	<title>Automation, Vol. 7, Pages 176: Adaptive Sliding Mode Position Control and Impedance Force Feedback for Bilateral Teleoperation</title>
	<link>https://www.mdpi.com/2673-4052/7/6/176</link>
	<description>This paper investigates bilateral teleoperation under time-varying communication delays and dynamic uncertainties. A role-specific asymmetric master–slave control architecture is proposed, in which impedance control is assigned to the master side for interaction-dynamics shaping and force feedback, while adaptive sliding mode control (SMC) is employed at the slave side for robust position tracking. A radial basis function neural network (RBFNN) is integrated into the slave-side SMC to compensate for uncertain nonlinear dynamics without requiring exact identification of the slave model. The adaptive laws are derived using Lyapunov stability analysis, and the approximation and adaptive estimation errors are explicitly considered. Under the stated boundedness conditions, the adaptive estimation errors and sliding variable remain bounded, and the slave position tracking error asymptotically converges to zero. The simulation results show that the proposed method substantially reduces the sensitivity of tracking performance to model uncertainty, with the IAPE remaining between 0.0031 and 0.0035 under 20–40% mismatch, compared with 0.2029–0.4057 for conventional SMC. Additional simulations with time-varying communication delays demonstrate bounded tracking performance, although delay-induced phase shifts and transient oscillations remain. The results demonstrate the robustness of the proposed role-specific architecture under the considered uncertainty and communication conditions, while higher-DOF and experimental validation remain subjects for future work.</description>
	<pubDate>2026-11-01</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 176: Adaptive Sliding Mode Position Control and Impedance Force Feedback for Bilateral Teleoperation</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/176">doi: 10.3390/automation7060176</a></p>
	<p>Authors:
		Thi Tong
		Minh Ho
		Quang Tran
		Hong Nguyen
		Ba Nguyen
		Trong Trinh
		Minh Duong
		</p>
	<p>This paper investigates bilateral teleoperation under time-varying communication delays and dynamic uncertainties. A role-specific asymmetric master–slave control architecture is proposed, in which impedance control is assigned to the master side for interaction-dynamics shaping and force feedback, while adaptive sliding mode control (SMC) is employed at the slave side for robust position tracking. A radial basis function neural network (RBFNN) is integrated into the slave-side SMC to compensate for uncertain nonlinear dynamics without requiring exact identification of the slave model. The adaptive laws are derived using Lyapunov stability analysis, and the approximation and adaptive estimation errors are explicitly considered. Under the stated boundedness conditions, the adaptive estimation errors and sliding variable remain bounded, and the slave position tracking error asymptotically converges to zero. The simulation results show that the proposed method substantially reduces the sensitivity of tracking performance to model uncertainty, with the IAPE remaining between 0.0031 and 0.0035 under 20–40% mismatch, compared with 0.2029–0.4057 for conventional SMC. Additional simulations with time-varying communication delays demonstrate bounded tracking performance, although delay-induced phase shifts and transient oscillations remain. The results demonstrate the robustness of the proposed role-specific architecture under the considered uncertainty and communication conditions, while higher-DOF and experimental validation remain subjects for future work.</p>
	]]></content:encoded>

	<dc:title>Adaptive Sliding Mode Position Control and Impedance Force Feedback for Bilateral Teleoperation</dc:title>
			<dc:creator>Thi Tong</dc:creator>
			<dc:creator>Minh Ho</dc:creator>
			<dc:creator>Quang Tran</dc:creator>
			<dc:creator>Hong Nguyen</dc:creator>
			<dc:creator>Ba Nguyen</dc:creator>
			<dc:creator>Trong Trinh</dc:creator>
			<dc:creator>Minh Duong</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060176</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-11-01</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-11-01</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>176</prism:startingPage>
		<prism:doi>10.3390/automation7060176</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/176</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/175">

	<title>Automation, Vol. 7, Pages 175: Energy Management System for a Hybrid Solar-Wind Power Plant with Digital Twin Diagnostics and SCADA Monitoring</title>
	<link>https://www.mdpi.com/2673-4052/7/6/175</link>
	<description>This paper presents the development and numerical verification of a hierarchical rule-based energy management system (EMS) for a hybrid photovoltaic&amp;amp;ndash;wind power plant with a battery storage system, a digital twin and a SCADA monitoring subsystem. The simulation model is implemented in the MATLAB/Simulink environment and includes photovoltaic and wind generators, a battery, an electrical load, a grid connection and a hierarchical rule-based energy management system (EMS). The control algorithm is based on calculating the current energy imbalance, estimating the state of charge of the battery, and selecting one of four power distribution modes: charging, discharging, electricity export and electricity import. The power-balance calculation was checked at 51 EMS evaluation points, with a reported maximum absolute discrepancy and RMSE of 0 kW at the numerical precision of the available results. The integral residual across the selected system-level accounting boundary over 60 min was &amp;amp;minus;8.88 &amp;amp;times; 10&amp;amp;minus;16 kWh. Internal conversion and interconnection losses lie outside this boundary; therefore, these residuals indicate the algebraic consistency of the recorded boundary power flows and do not quantify physical component losses. The digital twin is implemented as a parallel calculation and diagnostic subsystem for the parallel recalculation of power balance and dynamic monitoring of the battery&amp;amp;rsquo;s state of charge. For SOC estimation, an RMSE of 0.003225, an MAE of 0.002411 and a maximum absolute error of 0.007687 were obtained. The SCADA subsystem provides centralised collection of ten power, control and diagnostic signals. The main contribution of this study is a modular simulation and verification framework integrating a hierarchical rule-based energy management system (EMS), a parallel computational Digital Twin subsystem, and SCADA signal collection. The architecture separates power-balance calculation, operating-mode selection, power dispatch, diagnostic signal generation, and monitoring, allowing individual functional blocks to be assessed separately. The reported numerical results characterize the consistency and operation of the developed architecture under the simulated conditions. The study does not introduce a new optimization-based control algorithm or claim experimental validation on a physical energy facility. The framework provides a basis for future comparative evaluation of alternative control strategies and further experimental validation.</description>
	<pubDate>2026-10-29</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 175: Energy Management System for a Hybrid Solar-Wind Power Plant with Digital Twin Diagnostics and SCADA Monitoring</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/175">doi: 10.3390/automation7060175</a></p>
	<p>Authors:
		Baglan Bekbauov
		Aizhan Apsemetova
		Kuttybek Arystanbayev
		Elmurod Bozorov
		</p>
	<p>This paper presents the development and numerical verification of a hierarchical rule-based energy management system (EMS) for a hybrid photovoltaic&amp;amp;ndash;wind power plant with a battery storage system, a digital twin and a SCADA monitoring subsystem. The simulation model is implemented in the MATLAB/Simulink environment and includes photovoltaic and wind generators, a battery, an electrical load, a grid connection and a hierarchical rule-based energy management system (EMS). The control algorithm is based on calculating the current energy imbalance, estimating the state of charge of the battery, and selecting one of four power distribution modes: charging, discharging, electricity export and electricity import. The power-balance calculation was checked at 51 EMS evaluation points, with a reported maximum absolute discrepancy and RMSE of 0 kW at the numerical precision of the available results. The integral residual across the selected system-level accounting boundary over 60 min was &amp;amp;minus;8.88 &amp;amp;times; 10&amp;amp;minus;16 kWh. Internal conversion and interconnection losses lie outside this boundary; therefore, these residuals indicate the algebraic consistency of the recorded boundary power flows and do not quantify physical component losses. The digital twin is implemented as a parallel calculation and diagnostic subsystem for the parallel recalculation of power balance and dynamic monitoring of the battery&amp;amp;rsquo;s state of charge. For SOC estimation, an RMSE of 0.003225, an MAE of 0.002411 and a maximum absolute error of 0.007687 were obtained. The SCADA subsystem provides centralised collection of ten power, control and diagnostic signals. The main contribution of this study is a modular simulation and verification framework integrating a hierarchical rule-based energy management system (EMS), a parallel computational Digital Twin subsystem, and SCADA signal collection. The architecture separates power-balance calculation, operating-mode selection, power dispatch, diagnostic signal generation, and monitoring, allowing individual functional blocks to be assessed separately. The reported numerical results characterize the consistency and operation of the developed architecture under the simulated conditions. The study does not introduce a new optimization-based control algorithm or claim experimental validation on a physical energy facility. The framework provides a basis for future comparative evaluation of alternative control strategies and further experimental validation.</p>
	]]></content:encoded>

	<dc:title>Energy Management System for a Hybrid Solar-Wind Power Plant with Digital Twin Diagnostics and SCADA Monitoring</dc:title>
			<dc:creator>Baglan Bekbauov</dc:creator>
			<dc:creator>Aizhan Apsemetova</dc:creator>
			<dc:creator>Kuttybek Arystanbayev</dc:creator>
			<dc:creator>Elmurod Bozorov</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060175</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-29</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-29</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>175</prism:startingPage>
		<prism:doi>10.3390/automation7060175</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/175</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/174">

	<title>Automation, Vol. 7, Pages 174: Disturbance Rejection Technique on an Industrial Vacuum Furnace: The Model-Free iPID-DOB Approach</title>
	<link>https://www.mdpi.com/2673-4052/7/6/174</link>
	<description>Slow non-minimum-phase (NMP) processes present significant control challenges due to inverse response behaviour, sluggish dynamics, and sensitivity to disturbances and model uncertainty. This study investigates the performance of two model-independent control strategies, namely a disturbance observer-based intelligent proportional&amp;amp;ndash;integral controller (iPID-DOB) and a Particle Swarm Optimization-tuned Active Disturbance Rejection Controller (ADRC-PSO), for the regulation of a slow NMP process identified from experimental open-loop data of an industrial vacuum furnace. The iPID-DOB integrates a conventional PI structure with an online disturbance observer to estimate and compensate disturbance-related effects, whereas the ADRC employs an extended state observer (ESO) optimized using Particle Swarm Optimization to estimate system states and lumped uncertainties in real time. Both controllers were evaluated under identical conditions involving multiple reference changes, actuator constraints, and repeated disturbance scenarios. Performance was assessed using steady-state error, root-mean-square error (RMSE), rise time, settling time, overshoot, integral error indices (ISE, IAE, and ITAE), control effort, disturbance-rejection capability, and recovery time. The results demonstrate that both controllers achieved stable closed-loop operation and effective disturbance attenuation across the investigated operating range. However, the iPID-DOB consistently outperformed the ADRC-PSO in all quantitative performance metrics, exhibiting faster convergence, lower tracking error, improved disturbance rejection, and reduced recovery times. Conversely, the ADRC-PSO generated a significantly smoother and more stable control action, primarily due to the optimized extended state observer, which provided accurate estimation of process states and lumped disturbances while suppressing excessive control activity. The findings indicate that the iPID-DOB offers superior tracking and robustness performance for the considered slow NMP process, whereas the ADRC-PSO provides enhanced control smoothness and operational stability. These results highlight the trade-off between aggressive disturbance-compensation strategies and observer-driven robust control, providing practical guidance for selecting suitable model-independent control architectures for challenging industrial processes.</description>
	<pubDate>2026-10-28</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 174: Disturbance Rejection Technique on an Industrial Vacuum Furnace: The Model-Free iPID-DOB Approach</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/174">doi: 10.3390/automation7060174</a></p>
	<p>Authors:
		David N. Donkor
		Kingsley A. Ogudo
		Vikash Rameshar
		</p>
	<p>Slow non-minimum-phase (NMP) processes present significant control challenges due to inverse response behaviour, sluggish dynamics, and sensitivity to disturbances and model uncertainty. This study investigates the performance of two model-independent control strategies, namely a disturbance observer-based intelligent proportional&amp;amp;ndash;integral controller (iPID-DOB) and a Particle Swarm Optimization-tuned Active Disturbance Rejection Controller (ADRC-PSO), for the regulation of a slow NMP process identified from experimental open-loop data of an industrial vacuum furnace. The iPID-DOB integrates a conventional PI structure with an online disturbance observer to estimate and compensate disturbance-related effects, whereas the ADRC employs an extended state observer (ESO) optimized using Particle Swarm Optimization to estimate system states and lumped uncertainties in real time. Both controllers were evaluated under identical conditions involving multiple reference changes, actuator constraints, and repeated disturbance scenarios. Performance was assessed using steady-state error, root-mean-square error (RMSE), rise time, settling time, overshoot, integral error indices (ISE, IAE, and ITAE), control effort, disturbance-rejection capability, and recovery time. The results demonstrate that both controllers achieved stable closed-loop operation and effective disturbance attenuation across the investigated operating range. However, the iPID-DOB consistently outperformed the ADRC-PSO in all quantitative performance metrics, exhibiting faster convergence, lower tracking error, improved disturbance rejection, and reduced recovery times. Conversely, the ADRC-PSO generated a significantly smoother and more stable control action, primarily due to the optimized extended state observer, which provided accurate estimation of process states and lumped disturbances while suppressing excessive control activity. The findings indicate that the iPID-DOB offers superior tracking and robustness performance for the considered slow NMP process, whereas the ADRC-PSO provides enhanced control smoothness and operational stability. These results highlight the trade-off between aggressive disturbance-compensation strategies and observer-driven robust control, providing practical guidance for selecting suitable model-independent control architectures for challenging industrial processes.</p>
	]]></content:encoded>

	<dc:title>Disturbance Rejection Technique on an Industrial Vacuum Furnace: The Model-Free iPID-DOB Approach</dc:title>
			<dc:creator>David N. Donkor</dc:creator>
			<dc:creator>Kingsley A. Ogudo</dc:creator>
			<dc:creator>Vikash Rameshar</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060174</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-28</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-28</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>174</prism:startingPage>
		<prism:doi>10.3390/automation7060174</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/174</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/173">

	<title>Automation, Vol. 7, Pages 173: Sensitivity-Partitioned Switching Control of Pole-Mounted Capacitor-Reactor Devices in PV-Rich Distribution Feeders</title>
	<link>https://www.mdpi.com/2673-4052/7/6/173</link>
	<description>High photovoltaic (PV) penetration can cause bidirectional voltage violations in distribution feeders, with evening undervoltage and midday overvoltage requiring opposite reactive power responses. This paper proposes a sensitivity-partitioned switching control method for pole-mounted capacitor-reactor devices. A voltage-dependent signed discrete feasible region is formulated to represent capacitive and inductive operating states subject to mechanical interlocking. Voltage-reactive power sensitivity fingerprints are used to partition devices into control zones, in which finite-state switching candidates are screened locally and subsequently verified using nonlinear forward&amp;amp;ndash;backward sweep power flow. An operation count penalty is incorporated to reduce mechanical switching. The analysis starts with a fixed-compensation, capacitor-only setup as the baseline, then switches to a full capacitor-reactor controller with both positive and negative adjustment capability on a balanced IEEE 33-bus feeder with 160% PV penetration. The uncompensated feeder exhibits 16.00% undervoltage and 4.01% overvoltage violations. Sensitivity partitioning reduces the aggregate screening space from 27,000 full-feeder combinations to 215 zonal candidates and the mean per-step computation time from 7405 ms to 89 ms. The proposed controller eliminates both types of voltage violations with 14 daily switching actions, compared with 33 for local threshold control. The capacitor-only ablation retains 4.01% overvoltage violations, confirming the necessity of inductive reactive power absorption for the considered high-PV operating condition.</description>
	<pubDate>2026-10-27</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 173: Sensitivity-Partitioned Switching Control of Pole-Mounted Capacitor-Reactor Devices in PV-Rich Distribution Feeders</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/173">doi: 10.3390/automation7060173</a></p>
	<p>Authors:
		Jianhao Xiong
		Wei Zeng
		Haoming Wang
		Jiaoxin Jia
		Min Sun
		Hao He
		Zijing Wan
		Yan Zhang
		</p>
	<p>High photovoltaic (PV) penetration can cause bidirectional voltage violations in distribution feeders, with evening undervoltage and midday overvoltage requiring opposite reactive power responses. This paper proposes a sensitivity-partitioned switching control method for pole-mounted capacitor-reactor devices. A voltage-dependent signed discrete feasible region is formulated to represent capacitive and inductive operating states subject to mechanical interlocking. Voltage-reactive power sensitivity fingerprints are used to partition devices into control zones, in which finite-state switching candidates are screened locally and subsequently verified using nonlinear forward&amp;amp;ndash;backward sweep power flow. An operation count penalty is incorporated to reduce mechanical switching. The analysis starts with a fixed-compensation, capacitor-only setup as the baseline, then switches to a full capacitor-reactor controller with both positive and negative adjustment capability on a balanced IEEE 33-bus feeder with 160% PV penetration. The uncompensated feeder exhibits 16.00% undervoltage and 4.01% overvoltage violations. Sensitivity partitioning reduces the aggregate screening space from 27,000 full-feeder combinations to 215 zonal candidates and the mean per-step computation time from 7405 ms to 89 ms. The proposed controller eliminates both types of voltage violations with 14 daily switching actions, compared with 33 for local threshold control. The capacitor-only ablation retains 4.01% overvoltage violations, confirming the necessity of inductive reactive power absorption for the considered high-PV operating condition.</p>
	]]></content:encoded>

	<dc:title>Sensitivity-Partitioned Switching Control of Pole-Mounted Capacitor-Reactor Devices in PV-Rich Distribution Feeders</dc:title>
			<dc:creator>Jianhao Xiong</dc:creator>
			<dc:creator>Wei Zeng</dc:creator>
			<dc:creator>Haoming Wang</dc:creator>
			<dc:creator>Jiaoxin Jia</dc:creator>
			<dc:creator>Min Sun</dc:creator>
			<dc:creator>Hao He</dc:creator>
			<dc:creator>Zijing Wan</dc:creator>
			<dc:creator>Yan Zhang</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060173</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-27</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-27</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>173</prism:startingPage>
		<prism:doi>10.3390/automation7060173</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/173</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/172">

	<title>Automation, Vol. 7, Pages 172: A Stage-Based Framework for Embedded Implementation and Experimental Validation of Fractional-Order PID Controllers</title>
	<link>https://www.mdpi.com/2673-4052/7/6/172</link>
	<description>Fractional-order proportional&amp;amp;ndash;integral&amp;amp;ndash;derivative (FOPID) controllers provide additional tuning flexibility, but their operators require finite-dimensional discrete representations for implementation on embedded hardware. This study presents a stage-based workflow integrating plant identification, controller tuning, rational approximation, discretization, canonical infinite impulse response (IIR) realization, embedded implementation, and experimental evaluation. The workflow was demonstrated for Maxon DC motor speed control on an ATSAM3X8E ARM Cortex-M3 microcontroller. A FOPID controller tuned using the Ziegler&amp;amp;ndash;Nichols/Nelder&amp;amp;ndash;Mead procedure with the integral of time-weighted absolute error (ITAE) criterion was approximated by CRONE over 0.1&amp;amp;ndash;1000 rad/s and discretized by zero-order hold (ZOH) with a 1 ms sampling period. The resulting fifth-order discrete transfer function was implemented as a canonical IIR difference equation. Continuous- and discrete-time simulations produced similar transient responses. At a 2000 rpm motor-shaft reference (estimated 10 rpm gear-reducer output), the experiment yielded 2.0% overshoot, an integral absolute error of 0.00869, and an ITAE of 0.02249. Under the &amp;amp;plusmn;2% criterion, the response left the settling band after the first entry at 1.14 s, exceeding the 0.10 s target. These results support workflow feasibility while exposing a simulation-to-experiment gap due to unmodeled dynamics, measurement limitations, and implementation effects.</description>
	<pubDate>2026-10-26</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 172: A Stage-Based Framework for Embedded Implementation and Experimental Validation of Fractional-Order PID Controllers</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/172">doi: 10.3390/automation7060172</a></p>
	<p>Authors:
		Carlos Calderon-Cordova
		Roger Sarango
		Francklin Rivas-Echeverría
		</p>
	<p>Fractional-order proportional&amp;amp;ndash;integral&amp;amp;ndash;derivative (FOPID) controllers provide additional tuning flexibility, but their operators require finite-dimensional discrete representations for implementation on embedded hardware. This study presents a stage-based workflow integrating plant identification, controller tuning, rational approximation, discretization, canonical infinite impulse response (IIR) realization, embedded implementation, and experimental evaluation. The workflow was demonstrated for Maxon DC motor speed control on an ATSAM3X8E ARM Cortex-M3 microcontroller. A FOPID controller tuned using the Ziegler&amp;amp;ndash;Nichols/Nelder&amp;amp;ndash;Mead procedure with the integral of time-weighted absolute error (ITAE) criterion was approximated by CRONE over 0.1&amp;amp;ndash;1000 rad/s and discretized by zero-order hold (ZOH) with a 1 ms sampling period. The resulting fifth-order discrete transfer function was implemented as a canonical IIR difference equation. Continuous- and discrete-time simulations produced similar transient responses. At a 2000 rpm motor-shaft reference (estimated 10 rpm gear-reducer output), the experiment yielded 2.0% overshoot, an integral absolute error of 0.00869, and an ITAE of 0.02249. Under the &amp;amp;plusmn;2% criterion, the response left the settling band after the first entry at 1.14 s, exceeding the 0.10 s target. These results support workflow feasibility while exposing a simulation-to-experiment gap due to unmodeled dynamics, measurement limitations, and implementation effects.</p>
	]]></content:encoded>

	<dc:title>A Stage-Based Framework for Embedded Implementation and Experimental Validation of Fractional-Order PID Controllers</dc:title>
			<dc:creator>Carlos Calderon-Cordova</dc:creator>
			<dc:creator>Roger Sarango</dc:creator>
			<dc:creator>Francklin Rivas-Echeverría</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060172</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-26</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-26</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>172</prism:startingPage>
		<prism:doi>10.3390/automation7060172</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/172</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/171">

	<title>Automation, Vol. 7, Pages 171: Grey Wolf Optimization-Based Energy Management of BESS for Seamless Grid-Following Operation in PV-Integrated Microgrids</title>
	<link>https://www.mdpi.com/2673-4052/7/6/171</link>
	<description>The large-scale adoption of Photo Voltaic (PV) generation in today&amp;amp;rsquo;s microgrids poses substantial operational challenges, given the intermittent and uncertain nature of energy resources. These challenges result in power imbalances, voltage and frequency fluctuations, and degraded system resilience to grid disturbances. This paper proposes an intelligent energy management framework and control based on Grey Wolf Optimization for microgrids that incorporate battery energy storage systems. The proposed work effectively synchronizes the real-time PV generation, load demand, battery state of charge, and grid availability to attain adaptive power management in both grid-connected and island conditions. Differently from traditional control methods, this method allows seamless transition from Grid-Following (GFL) to Grid-Forming mode (GFM), and provides active and reactive power regulation, adaptive solar power curtailment, frequency stabilization, grid re-synchronization, and power imbalance compensation. This results in the effective control of the BESS, and thus, dynamic energy balancing, better utilization of renewable energy and voltage and frequency stability in transient operating conditions. The proposed control mechanism has been validated through extensive MATLAB simulations across various operational scenarios, which included changes in irradiance, fluctuations in load, renewable power containment, grid disconnection, island operation, and reconnection to the utility grid. Results from the simulations indicate that the GWO-based framework provides a quicker dynamic response, minimizes frequency deviations, increases renewable utilization, and exhibits a greater operational resilience when compared to conventional methods. The comparative simulation results demonstrate that the proposed GWO-based framework achieves a settling time, peak overshoot, solar-curtailment reduction, renewable energy utilization, and energy-harvest efficiency that outperforms the considered benchmark approaches under the investigated operating conditions. Additionally, the proposed strategy significantly improves the precision of energy sharing and ensures reliable operation of the microgrid (&amp;amp;micro;-grid) during more transitions and disturbances in the grid. These findings show that the GWO-based BESS energy management framework is an effective, robust, and scalable operation for improving the flexibility, stability, and reliability of next-generation PV-integrated microgrids characterized by high levels of renewable energy penetration.</description>
	<pubDate>2026-10-26</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 171: Grey Wolf Optimization-Based Energy Management of BESS for Seamless Grid-Following Operation in PV-Integrated Microgrids</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/171">doi: 10.3390/automation7060171</a></p>
	<p>Authors:
		M. Bala Siva Prasad
		P. Balachennaiah
		</p>
	<p>The large-scale adoption of Photo Voltaic (PV) generation in today&amp;amp;rsquo;s microgrids poses substantial operational challenges, given the intermittent and uncertain nature of energy resources. These challenges result in power imbalances, voltage and frequency fluctuations, and degraded system resilience to grid disturbances. This paper proposes an intelligent energy management framework and control based on Grey Wolf Optimization for microgrids that incorporate battery energy storage systems. The proposed work effectively synchronizes the real-time PV generation, load demand, battery state of charge, and grid availability to attain adaptive power management in both grid-connected and island conditions. Differently from traditional control methods, this method allows seamless transition from Grid-Following (GFL) to Grid-Forming mode (GFM), and provides active and reactive power regulation, adaptive solar power curtailment, frequency stabilization, grid re-synchronization, and power imbalance compensation. This results in the effective control of the BESS, and thus, dynamic energy balancing, better utilization of renewable energy and voltage and frequency stability in transient operating conditions. The proposed control mechanism has been validated through extensive MATLAB simulations across various operational scenarios, which included changes in irradiance, fluctuations in load, renewable power containment, grid disconnection, island operation, and reconnection to the utility grid. Results from the simulations indicate that the GWO-based framework provides a quicker dynamic response, minimizes frequency deviations, increases renewable utilization, and exhibits a greater operational resilience when compared to conventional methods. The comparative simulation results demonstrate that the proposed GWO-based framework achieves a settling time, peak overshoot, solar-curtailment reduction, renewable energy utilization, and energy-harvest efficiency that outperforms the considered benchmark approaches under the investigated operating conditions. Additionally, the proposed strategy significantly improves the precision of energy sharing and ensures reliable operation of the microgrid (&amp;amp;micro;-grid) during more transitions and disturbances in the grid. These findings show that the GWO-based BESS energy management framework is an effective, robust, and scalable operation for improving the flexibility, stability, and reliability of next-generation PV-integrated microgrids characterized by high levels of renewable energy penetration.</p>
	]]></content:encoded>

	<dc:title>Grey Wolf Optimization-Based Energy Management of BESS for Seamless Grid-Following Operation in PV-Integrated Microgrids</dc:title>
			<dc:creator>M. Bala Siva Prasad</dc:creator>
			<dc:creator>P. Balachennaiah</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060171</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-26</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-26</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>171</prism:startingPage>
		<prism:doi>10.3390/automation7060171</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/171</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/6/170">

	<title>Automation, Vol. 7, Pages 170: Contingent Planning in Domains with Unsafe Facts</title>
	<link>https://www.mdpi.com/2673-4052/7/6/170</link>
	<description>Contingent planning problems model agents with partial information about their state, that can use sensing actions to observe and reason about unknown values. Such problems can be efficiently solved using an online approach, where the agent replans each time new information is obtained through sensing. In this paper, we study Unsafe PPOS (U-PPOS) problems, where the input specification may contain unsafe facts&amp;amp;mdash;facts whose assumed value in the model differs from reality. Specifically, we consider false positives (FPs), which are assumed true but are actually false, and false negatives (FNs), which are assumed false but are actually true. FPs may lead planners to action failures, while FNs may cause the agent to believe that deadends exist. We propose methods to detect and handle both types of inaccuracies, establish their theoretical properties, and evaluate their performance across a set of benchmark domains.</description>
	<pubDate>2026-10-24</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 170: Contingent Planning in Domains with Unsafe Facts</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/6/170">doi: 10.3390/automation7060170</a></p>
	<p>Authors:
		Tomer Ravkaie
		Guy Shani
		</p>
	<p>Contingent planning problems model agents with partial information about their state, that can use sensing actions to observe and reason about unknown values. Such problems can be efficiently solved using an online approach, where the agent replans each time new information is obtained through sensing. In this paper, we study Unsafe PPOS (U-PPOS) problems, where the input specification may contain unsafe facts&amp;amp;mdash;facts whose assumed value in the model differs from reality. Specifically, we consider false positives (FPs), which are assumed true but are actually false, and false negatives (FNs), which are assumed false but are actually true. FPs may lead planners to action failures, while FNs may cause the agent to believe that deadends exist. We propose methods to detect and handle both types of inaccuracies, establish their theoretical properties, and evaluate their performance across a set of benchmark domains.</p>
	]]></content:encoded>

	<dc:title>Contingent Planning in Domains with Unsafe Facts</dc:title>
			<dc:creator>Tomer Ravkaie</dc:creator>
			<dc:creator>Guy Shani</dc:creator>
		<dc:identifier>doi: 10.3390/automation7060170</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-24</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-24</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>6</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>170</prism:startingPage>
		<prism:doi>10.3390/automation7060170</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/6/170</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/169">

	<title>Automation, Vol. 7, Pages 169: Requirements Engineering with Generative AI: A Systematic Literature Review of Using AI Agents for Requirements Engineering Automation</title>
	<link>https://www.mdpi.com/2673-4052/7/5/169</link>
	<description>Requirements development is a critical phase of software development, as the requirements developed during this phase provide the foundation for subsequent project activities. Because software requirements are predominantly expressed in natural language, automation of requirements engineering (RE) activities has historically been challenging. Recent advances in the natural language processing capabilities of Large Language Models (LLMs), however, have created new opportunities for automating RE tasks through AI agents. Accordingly, research on LLM-supported, agent-based RE automation has grown substantially in recent years. This study presents the results of a literature review of research on LLM-supported agent-based automation of software requirements engineering. The review identified 50 studies, of which 24 propose or describe agent-based architectures for RE automation, while the remaining 26 introduce benchmarks, baselines, evaluation environments, or comparative approaches for assessing agent-based solutions. The 24 studies describing an agent-based architecture are classified according to the requirements activities defined in the Software Requirements knowledge area of SWEBOK V4.0a, namely requirements elicitation, analysis, specification, validation, and tracing. Our analysis identifies the dominant agent architectures employed across RE activities, recurring design patterns, evaluation approaches, and key limitations of agent-based RE systems. Although significant advances have been made, several critical challenges remain, including hallucinations, scalability, and the lack of standardized evaluation frameworks. These findings highlight important directions for future research toward reliable, scalable, and systematically evaluated AI agent-based automation of software requirements engineering.</description>
	<pubDate>2026-10-21</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 169: Requirements Engineering with Generative AI: A Systematic Literature Review of Using AI Agents for Requirements Engineering Automation</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/169">doi: 10.3390/automation7050169</a></p>
	<p>Authors:
		Niraali Deepak Bandi
		Kadir Alpaslan Demir
		</p>
	<p>Requirements development is a critical phase of software development, as the requirements developed during this phase provide the foundation for subsequent project activities. Because software requirements are predominantly expressed in natural language, automation of requirements engineering (RE) activities has historically been challenging. Recent advances in the natural language processing capabilities of Large Language Models (LLMs), however, have created new opportunities for automating RE tasks through AI agents. Accordingly, research on LLM-supported, agent-based RE automation has grown substantially in recent years. This study presents the results of a literature review of research on LLM-supported agent-based automation of software requirements engineering. The review identified 50 studies, of which 24 propose or describe agent-based architectures for RE automation, while the remaining 26 introduce benchmarks, baselines, evaluation environments, or comparative approaches for assessing agent-based solutions. The 24 studies describing an agent-based architecture are classified according to the requirements activities defined in the Software Requirements knowledge area of SWEBOK V4.0a, namely requirements elicitation, analysis, specification, validation, and tracing. Our analysis identifies the dominant agent architectures employed across RE activities, recurring design patterns, evaluation approaches, and key limitations of agent-based RE systems. Although significant advances have been made, several critical challenges remain, including hallucinations, scalability, and the lack of standardized evaluation frameworks. These findings highlight important directions for future research toward reliable, scalable, and systematically evaluated AI agent-based automation of software requirements engineering.</p>
	]]></content:encoded>

	<dc:title>Requirements Engineering with Generative AI: A Systematic Literature Review of Using AI Agents for Requirements Engineering Automation</dc:title>
			<dc:creator>Niraali Deepak Bandi</dc:creator>
			<dc:creator>Kadir Alpaslan Demir</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050169</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-21</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-21</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Systematic Review</prism:section>
	<prism:startingPage>169</prism:startingPage>
		<prism:doi>10.3390/automation7050169</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/169</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/168">

	<title>Automation, Vol. 7, Pages 168: Robust and Adaptive Optimization of Carrier-Based Aircraft Operations Scheduling Considering Resource Degradation and Uncertainty</title>
	<link>https://www.mdpi.com/2673-4052/7/5/168</link>
	<description>To address the issue of scheduling failure in carrier-based aircraft support operations under complex sea conditions caused by resource performance degradation and stochastic environmental disturbances, this paper proposes a robust and adaptive scheduling optimization framework. First, a Gamma stochastic process is introduced to explicitly model the dynamic degradation evolution path of support resources, and a non-linear mapping relationship coupling equipment status with failure probability is constructed. Second, the scheduling problem is formulated as a finite-horizon Markov Decision Process (MDP), and a deep reinforcement learning solution scheme based on the Proximal Policy Optimization (PPO) algorithm is designed. To enhance system survivability under extreme uncertainties, a scenario-based robust training mechanism is developed, enabling the policy to cover multiple degradation trajectories. Finally, a rolling-horizon rescheduling mechanism is incorporated to achieve adaptive feedback adjustment to real-time status deviations. Simulation results demonstrate that the proposed method significantly reduces the total support duration (makespan) while effectively enhancing equipment reliability and the disturbance resistance of scheduling plans.</description>
	<pubDate>2026-10-20</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 168: Robust and Adaptive Optimization of Carrier-Based Aircraft Operations Scheduling Considering Resource Degradation and Uncertainty</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/168">doi: 10.3390/automation7050168</a></p>
	<p>Authors:
		Jian Yin
		Longyan Ding
		Liran Shen
		Xinwei Wang
		Yunsheng Fan
		</p>
	<p>To address the issue of scheduling failure in carrier-based aircraft support operations under complex sea conditions caused by resource performance degradation and stochastic environmental disturbances, this paper proposes a robust and adaptive scheduling optimization framework. First, a Gamma stochastic process is introduced to explicitly model the dynamic degradation evolution path of support resources, and a non-linear mapping relationship coupling equipment status with failure probability is constructed. Second, the scheduling problem is formulated as a finite-horizon Markov Decision Process (MDP), and a deep reinforcement learning solution scheme based on the Proximal Policy Optimization (PPO) algorithm is designed. To enhance system survivability under extreme uncertainties, a scenario-based robust training mechanism is developed, enabling the policy to cover multiple degradation trajectories. Finally, a rolling-horizon rescheduling mechanism is incorporated to achieve adaptive feedback adjustment to real-time status deviations. Simulation results demonstrate that the proposed method significantly reduces the total support duration (makespan) while effectively enhancing equipment reliability and the disturbance resistance of scheduling plans.</p>
	]]></content:encoded>

	<dc:title>Robust and Adaptive Optimization of Carrier-Based Aircraft Operations Scheduling Considering Resource Degradation and Uncertainty</dc:title>
			<dc:creator>Jian Yin</dc:creator>
			<dc:creator>Longyan Ding</dc:creator>
			<dc:creator>Liran Shen</dc:creator>
			<dc:creator>Xinwei Wang</dc:creator>
			<dc:creator>Yunsheng Fan</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050168</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-20</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-20</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>168</prism:startingPage>
		<prism:doi>10.3390/automation7050168</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/168</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/167">

	<title>Automation, Vol. 7, Pages 167: Experimental Evaluation of a Speed-Adaptive CBF-Inspired Steering Filter for Low-Speed Ackermann Navigation</title>
	<link>https://www.mdpi.com/2673-4052/7/5/167</link>
	<description>This study evaluates a speed-adaptive CBF-inspired steering filter for low-speed Ackermann navigation. A scalar quadratic program (QP) acts as a modular steering filter downstream of the nominal path tracker, while measured vehicle speed scales an adaptive elliptical clearance boundary. Simulation studies use Pure Pursuit and MPC nominal steering controllers, whereas the Quanser QCar uses PI speed regulation with Stanley steering. The constraint is inspired by control barrier function (CBF) filtering and is implemented through a reduced steering-channel control-affine approximation. Filter parameters are tuned in a MATLAB R2024bdigital twin and evaluated on the QCar in a structured indoor environment with static obstacles. In simulation, the filtered configurations avoided the collision observed with the unfiltered Pure Pursuit baseline, at the cost of increased path-tracking error. Five repeated QCar trials at 0.4&amp;amp;nbsp;m/s completed without collision and produced a mean pairwise dynamic-time-warping distance of 0.070&amp;amp;plusmn;0.030&amp;amp;nbsp;m. Across 15 fixed-obstacle QCar experiments, using rectangular footprints of 0.48&amp;amp;nbsp;m&amp;amp;times;0.21&amp;amp;nbsp;m for the vehicle and 0.20&amp;amp;nbsp;m&amp;amp;times;0.08&amp;amp;nbsp;m for the obstacle, the estimated minimum surface-to-surface clearance was 0.274&amp;amp;plusmn;0.041&amp;amp;nbsp;m (mean &amp;amp;plusmn; sample SD), with a 95% confidence interval for the mean of 0.251&amp;amp;ndash;0.296&amp;amp;nbsp;m and an observed range of 0.209&amp;amp;ndash;0.336&amp;amp;nbsp;m; all 15 trials completed without obstacle contact. The results establish empirical feasibility and quantify the simulated clearance&amp;amp;ndash;tracking trade-off within the stated low-speed, static-obstacle operating range.</description>
	<pubDate>2026-10-19</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 167: Experimental Evaluation of a Speed-Adaptive CBF-Inspired Steering Filter for Low-Speed Ackermann Navigation</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/167">doi: 10.3390/automation7050167</a></p>
	<p>Authors:
		Jesús O. Rodríguez-Ramírez
		Hugo G. Gonzalez-Hernandez
		Jorge A. Reyes-Avendaño
		Riccardo Cespi
		Stefano Di Gennaro
		Jose Martinez-Carranza
		</p>
	<p>This study evaluates a speed-adaptive CBF-inspired steering filter for low-speed Ackermann navigation. A scalar quadratic program (QP) acts as a modular steering filter downstream of the nominal path tracker, while measured vehicle speed scales an adaptive elliptical clearance boundary. Simulation studies use Pure Pursuit and MPC nominal steering controllers, whereas the Quanser QCar uses PI speed regulation with Stanley steering. The constraint is inspired by control barrier function (CBF) filtering and is implemented through a reduced steering-channel control-affine approximation. Filter parameters are tuned in a MATLAB R2024bdigital twin and evaluated on the QCar in a structured indoor environment with static obstacles. In simulation, the filtered configurations avoided the collision observed with the unfiltered Pure Pursuit baseline, at the cost of increased path-tracking error. Five repeated QCar trials at 0.4&amp;amp;nbsp;m/s completed without collision and produced a mean pairwise dynamic-time-warping distance of 0.070&amp;amp;plusmn;0.030&amp;amp;nbsp;m. Across 15 fixed-obstacle QCar experiments, using rectangular footprints of 0.48&amp;amp;nbsp;m&amp;amp;times;0.21&amp;amp;nbsp;m for the vehicle and 0.20&amp;amp;nbsp;m&amp;amp;times;0.08&amp;amp;nbsp;m for the obstacle, the estimated minimum surface-to-surface clearance was 0.274&amp;amp;plusmn;0.041&amp;amp;nbsp;m (mean &amp;amp;plusmn; sample SD), with a 95% confidence interval for the mean of 0.251&amp;amp;ndash;0.296&amp;amp;nbsp;m and an observed range of 0.209&amp;amp;ndash;0.336&amp;amp;nbsp;m; all 15 trials completed without obstacle contact. The results establish empirical feasibility and quantify the simulated clearance&amp;amp;ndash;tracking trade-off within the stated low-speed, static-obstacle operating range.</p>
	]]></content:encoded>

	<dc:title>Experimental Evaluation of a Speed-Adaptive CBF-Inspired Steering Filter for Low-Speed Ackermann Navigation</dc:title>
			<dc:creator>Jesús O. Rodríguez-Ramírez</dc:creator>
			<dc:creator>Hugo G. Gonzalez-Hernandez</dc:creator>
			<dc:creator>Jorge A. Reyes-Avendaño</dc:creator>
			<dc:creator>Riccardo Cespi</dc:creator>
			<dc:creator>Stefano Di Gennaro</dc:creator>
			<dc:creator>Jose Martinez-Carranza</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050167</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-19</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-19</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>167</prism:startingPage>
		<prism:doi>10.3390/automation7050167</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/167</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/165">

	<title>Automation, Vol. 7, Pages 165: Text-Guided Semantic Enhancement for Personal Protective Equipment Detection in Chemical Industry Environments</title>
	<link>https://www.mdpi.com/2673-4052/7/5/165</link>
	<description>The automated detection of personal protective equipment (PPE) is crucial for ensuring operational safety in hazardous environments. However, background interference and target semantic ambiguity caused by dramatic illumination changes and occlusions pose severe challenges to traditional deep learning models. To address these challenges, we present TGF-ChD, a detection model based on the text-guided fusion (TGF) module, which enhances feature representation by introducing category semantic priors. TGF-ChD first employs text-guided attention to spatially capture significant regions based on semantic constraints; subsequently, through semantic feature modulation, which calibrates channel gains and shifts according to semantic relevance, high-level semantic information is deeply embedded into multi-scale visual features. On the Chemical Industry PPE dataset, the model achieved an mAP of 54.3% and an mAP50 of 84.9%, exceeding TOOD by 2.0 and 6.0 percentage points, respectively. Furthermore, the model achieves improved performance on challenging categories, such as &amp;amp;lsquo;blurred clothing&amp;amp;rsquo; and &amp;amp;lsquo;blurred heads&amp;amp;rsquo; caused by severe reflections. These results indicate that jointly applying spatial and channel-level semantic conditioning can improve PPE detection under the evaluated complex industrial conditions.</description>
	<pubDate>2026-10-17</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 165: Text-Guided Semantic Enhancement for Personal Protective Equipment Detection in Chemical Industry Environments</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/165">doi: 10.3390/automation7050165</a></p>
	<p>Authors:
		Xuesheng Li
		Piaofang Yu
		Bo Lin
		</p>
	<p>The automated detection of personal protective equipment (PPE) is crucial for ensuring operational safety in hazardous environments. However, background interference and target semantic ambiguity caused by dramatic illumination changes and occlusions pose severe challenges to traditional deep learning models. To address these challenges, we present TGF-ChD, a detection model based on the text-guided fusion (TGF) module, which enhances feature representation by introducing category semantic priors. TGF-ChD first employs text-guided attention to spatially capture significant regions based on semantic constraints; subsequently, through semantic feature modulation, which calibrates channel gains and shifts according to semantic relevance, high-level semantic information is deeply embedded into multi-scale visual features. On the Chemical Industry PPE dataset, the model achieved an mAP of 54.3% and an mAP50 of 84.9%, exceeding TOOD by 2.0 and 6.0 percentage points, respectively. Furthermore, the model achieves improved performance on challenging categories, such as &amp;amp;lsquo;blurred clothing&amp;amp;rsquo; and &amp;amp;lsquo;blurred heads&amp;amp;rsquo; caused by severe reflections. These results indicate that jointly applying spatial and channel-level semantic conditioning can improve PPE detection under the evaluated complex industrial conditions.</p>
	]]></content:encoded>

	<dc:title>Text-Guided Semantic Enhancement for Personal Protective Equipment Detection in Chemical Industry Environments</dc:title>
			<dc:creator>Xuesheng Li</dc:creator>
			<dc:creator>Piaofang Yu</dc:creator>
			<dc:creator>Bo Lin</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050165</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-17</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-17</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>165</prism:startingPage>
		<prism:doi>10.3390/automation7050165</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/165</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/166">

	<title>Automation, Vol. 7, Pages 166: Distributed PI Observer-Based Sensor Fault Estimation in Networked Multi-Agent Systems with Application to Unmanned Ground Vehicles</title>
	<link>https://www.mdpi.com/2673-4052/7/5/166</link>
	<description>This paper addresses the problem of distributed sensor fault estimation (FE) in leader-following multi-agent systems (MASs) composed of unmanned ground vehicles (UGVs). In consensus-based formation control, sensor faults affect the local measurements exchanged among neighboring agents, degrading state estimation and compromising the tracking performance of the entire fleet. Although fault estimation techniques have been widely investigated, most existing distributed formulations do not explicitly address sensor FE within a leader-following consensus framework. A distributed Proportional&amp;amp;ndash;Integral (PI) observer is proposed to simultaneously reconstruct the agents&amp;amp;rsquo; states and estimate sensor fault signals using local and neighbor information. The observer and the leader-following consensus controller are considered within a unified design framework, and sufficient conditions for the simultaneous synthesis of the observer and consensus-controller gains are derived in terms of linear matrix inequalities (LMIs). The proposed methodology is validated through a cooperative UGV formation scenario, where simulation results demonstrate the ability of the distributed observer to reconstruct sensor fault signals and illustrate the effect of sensor faults on the leader-following tracking performance. The proposed approach focuses on sensor fault estimation and does not incorporate an active fault compensation mechanism.</description>
	<pubDate>2026-10-17</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 166: Distributed PI Observer-Based Sensor Fault Estimation in Networked Multi-Agent Systems with Application to Unmanned Ground Vehicles</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/166">doi: 10.3390/automation7050166</a></p>
	<p>Authors:
		Jesus A. Vazquez Trejo
		Arturo D. Espinosa Pedroza
		Emiliano Peña-Chang
		José F. Guerrero-Castellanos
		Manuel Adam-Medina
		Juan A. Vazquez Trejo
		</p>
	<p>This paper addresses the problem of distributed sensor fault estimation (FE) in leader-following multi-agent systems (MASs) composed of unmanned ground vehicles (UGVs). In consensus-based formation control, sensor faults affect the local measurements exchanged among neighboring agents, degrading state estimation and compromising the tracking performance of the entire fleet. Although fault estimation techniques have been widely investigated, most existing distributed formulations do not explicitly address sensor FE within a leader-following consensus framework. A distributed Proportional&amp;amp;ndash;Integral (PI) observer is proposed to simultaneously reconstruct the agents&amp;amp;rsquo; states and estimate sensor fault signals using local and neighbor information. The observer and the leader-following consensus controller are considered within a unified design framework, and sufficient conditions for the simultaneous synthesis of the observer and consensus-controller gains are derived in terms of linear matrix inequalities (LMIs). The proposed methodology is validated through a cooperative UGV formation scenario, where simulation results demonstrate the ability of the distributed observer to reconstruct sensor fault signals and illustrate the effect of sensor faults on the leader-following tracking performance. The proposed approach focuses on sensor fault estimation and does not incorporate an active fault compensation mechanism.</p>
	]]></content:encoded>

	<dc:title>Distributed PI Observer-Based Sensor Fault Estimation in Networked Multi-Agent Systems with Application to Unmanned Ground Vehicles</dc:title>
			<dc:creator>Jesus A. Vazquez Trejo</dc:creator>
			<dc:creator>Arturo D. Espinosa Pedroza</dc:creator>
			<dc:creator>Emiliano Peña-Chang</dc:creator>
			<dc:creator>José F. Guerrero-Castellanos</dc:creator>
			<dc:creator>Manuel Adam-Medina</dc:creator>
			<dc:creator>Juan A. Vazquez Trejo</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050166</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-17</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-17</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>166</prism:startingPage>
		<prism:doi>10.3390/automation7050166</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/166</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/164">

	<title>Automation, Vol. 7, Pages 164: Strict Passivity Analysis of UAV Swarm Cooperative Payload Transportation Systems with Time Delays</title>
	<link>https://www.mdpi.com/2673-4052/7/5/164</link>
	<description>This paper investigates strict passivity analysis for continuous-time unmanned aerial vehicle (UAV) swarm cooperative payload transportation systems with time delays arising from onboard processing and state acquisition. The UAVs are assumed to have identical dynamic characteristics and to be interconnected through a structurally sparse topology. By exploiting the block-diagonal structure of the subsystem matrices and the sparsity of the structural connectivity matrix (SCM), two necessary and sufficient strict passivity conditions that are mathematically equivalent to the conventional lumped criterion are derived without requiring high-dimensional matrix inversion. Furthermore, two sufficient conditions based on single-UAV parameters are developed to reduce the computational burden for large-scale swarms. Numerical studies based on UAV model parameters and generic subsystem matrices demonstrate the computational advantages of the proposed conditions. For a 14-UAV system, the conventional lumped criterion requires 2069.1 s, whereas the aggregate computation times of the two single-UAV-based conditions are 0.589 s and 0.336 s, respectively. These results demonstrate that the proposed structural decomposition framework can substantially improve the efficiency of strict passivity verification for large-scale UAV swarm systems.</description>
	<pubDate>2026-10-14</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 164: Strict Passivity Analysis of UAV Swarm Cooperative Payload Transportation Systems with Time Delays</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/164">doi: 10.3390/automation7050164</a></p>
	<p>Authors:
		Ziwen Wang
		Huabo Liu
		</p>
	<p>This paper investigates strict passivity analysis for continuous-time unmanned aerial vehicle (UAV) swarm cooperative payload transportation systems with time delays arising from onboard processing and state acquisition. The UAVs are assumed to have identical dynamic characteristics and to be interconnected through a structurally sparse topology. By exploiting the block-diagonal structure of the subsystem matrices and the sparsity of the structural connectivity matrix (SCM), two necessary and sufficient strict passivity conditions that are mathematically equivalent to the conventional lumped criterion are derived without requiring high-dimensional matrix inversion. Furthermore, two sufficient conditions based on single-UAV parameters are developed to reduce the computational burden for large-scale swarms. Numerical studies based on UAV model parameters and generic subsystem matrices demonstrate the computational advantages of the proposed conditions. For a 14-UAV system, the conventional lumped criterion requires 2069.1 s, whereas the aggregate computation times of the two single-UAV-based conditions are 0.589 s and 0.336 s, respectively. These results demonstrate that the proposed structural decomposition framework can substantially improve the efficiency of strict passivity verification for large-scale UAV swarm systems.</p>
	]]></content:encoded>

	<dc:title>Strict Passivity Analysis of UAV Swarm Cooperative Payload Transportation Systems with Time Delays</dc:title>
			<dc:creator>Ziwen Wang</dc:creator>
			<dc:creator>Huabo Liu</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050164</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-14</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-14</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>164</prism:startingPage>
		<prism:doi>10.3390/automation7050164</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/164</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/163">

	<title>Automation, Vol. 7, Pages 163: Electromagnetic Compatibility Design Framework for Wireless Railway Axle-Box Monitoring Systems</title>
	<link>https://www.mdpi.com/2673-4052/7/5/163</link>
	<description>Reliable wireless communication is a key requirement for railway axle-box condition monitoring, particularly in electromagnetically harsh environments where electromagnetic interference (EMI) may degrade diagnostic data transmission. This study proposes an Electromagnetic Compatibility Design Framework (EMCDF) for receiver-oriented assessment of electromagnetic exposure, equivalent EMC mitigation, and wireless-link performance. The framework integrates frequency-domain electromagnetic characterization with analytical radio-link, receiver-noise, residual-interference, and communication-performance models, establishing a physically traceable pathway from prescribed electromagnetic exposure to residual receiver-level interference and the resulting communication degradation. Prescribed in-band EMI scenarios and equivalent EMC-mitigation conditions are evaluated using the signal-to-interference-plus-noise ratio (SINR), theoretical channel capacity, and two proposed normalized electromagnetic robustness metrics based on SINR and theoretical capacity. Under the prescribed analytical scenarios, equivalent in-band EMC attenuation values of 15&amp;amp;ndash;20 dB correspond to interference-power reduction factors of approximately 31.6&amp;amp;ndash;100. For the adopted 433 MHz radio-link parameters, the model-derived SINR under the severe-EMI scenario decreases from approximately 37.83 dB at 1 km to 17.83 dB at 10 km, while the capacity-based robustness coefficient decreases from approximately 0.474 to 0.299 over the same distance range. The model-based results show that propagation loss and residual EMI influence receiver-level communication performance through physically distinct mechanisms and can therefore be evaluated separately within the proposed framework. The quantitative results represent comparative model-derived estimates rather than experimental validation over the investigated distance range. The proposed EMCDF provides a receiver-oriented analytical methodology for EMC-aware assessment and comparative evaluation of wireless railway condition-monitoring systems.</description>
	<pubDate>2026-10-14</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 163: Electromagnetic Compatibility Design Framework for Wireless Railway Axle-Box Monitoring Systems</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/163">doi: 10.3390/automation7050163</a></p>
	<p>Authors:
		Nuriddin N. Irgashev
		Janibek F. Kurbanov
		Asliddin Sh. Khurramov
		</p>
	<p>Reliable wireless communication is a key requirement for railway axle-box condition monitoring, particularly in electromagnetically harsh environments where electromagnetic interference (EMI) may degrade diagnostic data transmission. This study proposes an Electromagnetic Compatibility Design Framework (EMCDF) for receiver-oriented assessment of electromagnetic exposure, equivalent EMC mitigation, and wireless-link performance. The framework integrates frequency-domain electromagnetic characterization with analytical radio-link, receiver-noise, residual-interference, and communication-performance models, establishing a physically traceable pathway from prescribed electromagnetic exposure to residual receiver-level interference and the resulting communication degradation. Prescribed in-band EMI scenarios and equivalent EMC-mitigation conditions are evaluated using the signal-to-interference-plus-noise ratio (SINR), theoretical channel capacity, and two proposed normalized electromagnetic robustness metrics based on SINR and theoretical capacity. Under the prescribed analytical scenarios, equivalent in-band EMC attenuation values of 15&amp;amp;ndash;20 dB correspond to interference-power reduction factors of approximately 31.6&amp;amp;ndash;100. For the adopted 433 MHz radio-link parameters, the model-derived SINR under the severe-EMI scenario decreases from approximately 37.83 dB at 1 km to 17.83 dB at 10 km, while the capacity-based robustness coefficient decreases from approximately 0.474 to 0.299 over the same distance range. The model-based results show that propagation loss and residual EMI influence receiver-level communication performance through physically distinct mechanisms and can therefore be evaluated separately within the proposed framework. The quantitative results represent comparative model-derived estimates rather than experimental validation over the investigated distance range. The proposed EMCDF provides a receiver-oriented analytical methodology for EMC-aware assessment and comparative evaluation of wireless railway condition-monitoring systems.</p>
	]]></content:encoded>

	<dc:title>Electromagnetic Compatibility Design Framework for Wireless Railway Axle-Box Monitoring Systems</dc:title>
			<dc:creator>Nuriddin N. Irgashev</dc:creator>
			<dc:creator>Janibek F. Kurbanov</dc:creator>
			<dc:creator>Asliddin Sh. Khurramov</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050163</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-14</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-14</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>163</prism:startingPage>
		<prism:doi>10.3390/automation7050163</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/163</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/162">

	<title>Automation, Vol. 7, Pages 162: Actuation-Aware Reference Reshaping with Adaptive Super-Twisting Tracking for Heterogeneous UAV Formations During Obstacle-Induced Maneuvers</title>
	<link>https://www.mdpi.com/2673-4052/7/5/162</link>
	<description>Heterogeneous unmanned aerial vehicle (UAV) formations face competing tracking and coordination demands during obstacle-induced maneuvers with follower-specific acceleration limits. Formation-reshaping adaptive super-twisting control (FR-ASTWC) is developed using a leader&amp;amp;ndash;neighbor deformation cue to drive an anchored, acceleration-limited reference before a common tracking loop. Under identical plants and tuning, FR-ASTWC is compared with direct adaptive super-twisting control (D-ASTWC), coordination-input ASTWC (C-ASTWC), and coordination-filtered-input ASTWC (CF-ASTWC). For the nominal exact drag model, ideal D-ASTWC and FR-ASTWC retain the exogenous inner residual and satisfy a verified common quadratic certificate; C-ASTWC and CF-ASTWC are empirical ablations because their residual contains the cue. At 70% capability, FR-ASTWC reduces active peak deformation, mean deformation, and plant saturation exposure relative to CF-ASTWC by 64.9%, 48.2%, and 50.8%, respectively. Across two bypass directions and three disturbance-phase variants, the mean reductions are 59.3%, 45.0%, and 47.2%. A further 1200 simulations use 100 paired uncertainty draws at each capability level. At 70%, the mean paired reductions relative to CF-ASTWC are 53.8%, 43.4%, and 46.7% for the same three metrics. Parameter sweeps and the 85% and 100% cases expose reference return and low-demand trade-offs. The evidence therefore supports a conditional architectural advantage under active heterogeneous constraints, not universal superiority or a collision avoidance guarantee.</description>
	<pubDate>2026-10-12</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 162: Actuation-Aware Reference Reshaping with Adaptive Super-Twisting Tracking for Heterogeneous UAV Formations During Obstacle-Induced Maneuvers</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/162">doi: 10.3390/automation7050162</a></p>
	<p>Authors:
		Mohammed Ali Alfayizi
		</p>
	<p>Heterogeneous unmanned aerial vehicle (UAV) formations face competing tracking and coordination demands during obstacle-induced maneuvers with follower-specific acceleration limits. Formation-reshaping adaptive super-twisting control (FR-ASTWC) is developed using a leader&amp;amp;ndash;neighbor deformation cue to drive an anchored, acceleration-limited reference before a common tracking loop. Under identical plants and tuning, FR-ASTWC is compared with direct adaptive super-twisting control (D-ASTWC), coordination-input ASTWC (C-ASTWC), and coordination-filtered-input ASTWC (CF-ASTWC). For the nominal exact drag model, ideal D-ASTWC and FR-ASTWC retain the exogenous inner residual and satisfy a verified common quadratic certificate; C-ASTWC and CF-ASTWC are empirical ablations because their residual contains the cue. At 70% capability, FR-ASTWC reduces active peak deformation, mean deformation, and plant saturation exposure relative to CF-ASTWC by 64.9%, 48.2%, and 50.8%, respectively. Across two bypass directions and three disturbance-phase variants, the mean reductions are 59.3%, 45.0%, and 47.2%. A further 1200 simulations use 100 paired uncertainty draws at each capability level. At 70%, the mean paired reductions relative to CF-ASTWC are 53.8%, 43.4%, and 46.7% for the same three metrics. Parameter sweeps and the 85% and 100% cases expose reference return and low-demand trade-offs. The evidence therefore supports a conditional architectural advantage under active heterogeneous constraints, not universal superiority or a collision avoidance guarantee.</p>
	]]></content:encoded>

	<dc:title>Actuation-Aware Reference Reshaping with Adaptive Super-Twisting Tracking for Heterogeneous UAV Formations During Obstacle-Induced Maneuvers</dc:title>
			<dc:creator>Mohammed Ali Alfayizi</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050162</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-12</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-12</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>162</prism:startingPage>
		<prism:doi>10.3390/automation7050162</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/162</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/160">

	<title>Automation, Vol. 7, Pages 160: Development and Investigation of a Navigation Electronic Seal System with an Integrated Satellite Communication Channel</title>
	<link>https://www.mdpi.com/2673-4052/7/5/160</link>
	<description>This study presents a compact navigation electronic seal system employing the Iridium SBD9603 radio modem and a specialized antenna for transmitting Short Burst Data (SBD) messages. The primary objective was to integrate a satellite antenna within the constrained space of the device enclosure. Due to the impossibility of the manufacturer-recommended horizontal antenna placement, the antenna was installed at a 70&amp;amp;deg; angle. Experimental results confirmed the feasibility of this configuration. In open-space conditions, a stable signal level sufficient for communication was observed 88.6% of the time, while inside a protective enclosure made of PA12 + GF composite material, this value decreased to 65.3%. The average data packet delivery time ranged from 8 to 14 s. These results are explained by the dielectric properties of the enclosure material and partial signal shielding. At the same time, short-term signal fading (on average up to 43 s) was considered acceptable, since a typical SBD transaction takes less than 10 s, allowing efficient use of message retransmission. An important finding of the study is that when the seal is mounted on a steel container door, the inclined antenna performs more efficiently than the horizontal configuration (65.3% versus 61.1%). This effect is attributed to the metallic surface acting as a reflective screen, optimizing the electromagnetic conditions for the inclined antenna aperture. The obtained results can be applied in remote monitoring and identification systems for cargo transported in containers and trucks, including operation under challenging environmental conditions.</description>
	<pubDate>2026-10-08</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 160: Development and Investigation of a Navigation Electronic Seal System with an Integrated Satellite Communication Channel</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/160">doi: 10.3390/automation7050160</a></p>
	<p>Authors:
		Alexandr Inchin
		Aigul Kulakayeva
		Sabyrzhan Zhumagali
		Meirbek Moldabekov
		Dastan Yergaliyev
		</p>
	<p>This study presents a compact navigation electronic seal system employing the Iridium SBD9603 radio modem and a specialized antenna for transmitting Short Burst Data (SBD) messages. The primary objective was to integrate a satellite antenna within the constrained space of the device enclosure. Due to the impossibility of the manufacturer-recommended horizontal antenna placement, the antenna was installed at a 70&amp;amp;deg; angle. Experimental results confirmed the feasibility of this configuration. In open-space conditions, a stable signal level sufficient for communication was observed 88.6% of the time, while inside a protective enclosure made of PA12 + GF composite material, this value decreased to 65.3%. The average data packet delivery time ranged from 8 to 14 s. These results are explained by the dielectric properties of the enclosure material and partial signal shielding. At the same time, short-term signal fading (on average up to 43 s) was considered acceptable, since a typical SBD transaction takes less than 10 s, allowing efficient use of message retransmission. An important finding of the study is that when the seal is mounted on a steel container door, the inclined antenna performs more efficiently than the horizontal configuration (65.3% versus 61.1%). This effect is attributed to the metallic surface acting as a reflective screen, optimizing the electromagnetic conditions for the inclined antenna aperture. The obtained results can be applied in remote monitoring and identification systems for cargo transported in containers and trucks, including operation under challenging environmental conditions.</p>
	]]></content:encoded>

	<dc:title>Development and Investigation of a Navigation Electronic Seal System with an Integrated Satellite Communication Channel</dc:title>
			<dc:creator>Alexandr Inchin</dc:creator>
			<dc:creator>Aigul Kulakayeva</dc:creator>
			<dc:creator>Sabyrzhan Zhumagali</dc:creator>
			<dc:creator>Meirbek Moldabekov</dc:creator>
			<dc:creator>Dastan Yergaliyev</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050160</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-08</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-08</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>160</prism:startingPage>
		<prism:doi>10.3390/automation7050160</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/160</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/161">

	<title>Automation, Vol. 7, Pages 161: Reciprocal Zeroing Neurodynamics for Inverse-Free Manipulator Kinematic Control with Unknown Jacobian</title>
	<link>https://www.mdpi.com/2673-4052/7/5/161</link>
	<description>The kinematic control of manipulators has extensive applications in modern industry. Precise control enables robotic manipulators to perform diverse complex tasks, boost production efficiency/quality, and cut labor costs and safety risks. Kinematic control of manipulators centers on solving time-variant nonlinear equations. Traditional methods for this typically require a known Jacobian or matrix inversion. However, in practical applications, exact model parameters are typically unavailable, and computational resources are constrained. Moreover, the computational cost associated with matrix inverse is prohibitive. We propose an algorithm based on reciprocal zeroing neurodynamics for solving the time-variant nonlinear equation system of manipulator kinematic control with an unknown Jacobian matrix that is also inverse-free, which operates under the assumptions of continuous time formulation and perfect measurement conditions. We also provide theorems guaranteeing bounded convergence of Jacobian estimation error under noisy inputs. The proposed algorithm implicitly learns the unmeasurable kinematic model through online estimation of the Jacobian matrix based on real-time input&amp;amp;minus;output relationships and approximates the solutions of the nonlinear equation system, thereby enabling precise control of manipulator kinematics in an inverse-free manner. Furthermore, the effectiveness and correctness of the proposed algorithm are ensured through the rigorous stability and convergence guarantees established through Lyapunov-based analysis. These analyses are conducted using concepts, lemmas, theorems, and remarks, laying a solid foundation for evaluating the proposed algorithm&amp;amp;rsquo;s performance. Simulation-based and physical experiments exhibit tracking error constrained within the sub-millimeter to micrometer level, confirming the algorithm&amp;amp;rsquo;s efficacy and highlighting its robustness in handling unknown Jacobians without inverse matrix computation.</description>
	<pubDate>2026-10-08</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 161: Reciprocal Zeroing Neurodynamics for Inverse-Free Manipulator Kinematic Control with Unknown Jacobian</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/161">doi: 10.3390/automation7050161</a></p>
	<p>Authors:
		Yunong Zhang
		Dongqing Wu
		Shuai Li
		</p>
	<p>The kinematic control of manipulators has extensive applications in modern industry. Precise control enables robotic manipulators to perform diverse complex tasks, boost production efficiency/quality, and cut labor costs and safety risks. Kinematic control of manipulators centers on solving time-variant nonlinear equations. Traditional methods for this typically require a known Jacobian or matrix inversion. However, in practical applications, exact model parameters are typically unavailable, and computational resources are constrained. Moreover, the computational cost associated with matrix inverse is prohibitive. We propose an algorithm based on reciprocal zeroing neurodynamics for solving the time-variant nonlinear equation system of manipulator kinematic control with an unknown Jacobian matrix that is also inverse-free, which operates under the assumptions of continuous time formulation and perfect measurement conditions. We also provide theorems guaranteeing bounded convergence of Jacobian estimation error under noisy inputs. The proposed algorithm implicitly learns the unmeasurable kinematic model through online estimation of the Jacobian matrix based on real-time input&amp;amp;minus;output relationships and approximates the solutions of the nonlinear equation system, thereby enabling precise control of manipulator kinematics in an inverse-free manner. Furthermore, the effectiveness and correctness of the proposed algorithm are ensured through the rigorous stability and convergence guarantees established through Lyapunov-based analysis. These analyses are conducted using concepts, lemmas, theorems, and remarks, laying a solid foundation for evaluating the proposed algorithm&amp;amp;rsquo;s performance. Simulation-based and physical experiments exhibit tracking error constrained within the sub-millimeter to micrometer level, confirming the algorithm&amp;amp;rsquo;s efficacy and highlighting its robustness in handling unknown Jacobians without inverse matrix computation.</p>
	]]></content:encoded>

	<dc:title>Reciprocal Zeroing Neurodynamics for Inverse-Free Manipulator Kinematic Control with Unknown Jacobian</dc:title>
			<dc:creator>Yunong Zhang</dc:creator>
			<dc:creator>Dongqing Wu</dc:creator>
			<dc:creator>Shuai Li</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050161</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-08</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-08</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>161</prism:startingPage>
		<prism:doi>10.3390/automation7050161</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/161</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/159">

	<title>Automation, Vol. 7, Pages 159: EVRoute: An Open-Source Intelligent Routing and Decision-Support Platform for Electric Vehicles</title>
	<link>https://www.mdpi.com/2673-4052/7/5/159</link>
	<description>Electric vehicle (EV) adoption has increased the demand for intelligent trip-planning solutions capable of addressing charging infrastructure limitations and reducing range anxiety, a socio-technical barrier that disproportionately affects regions with developing charging networks. This paper presents EVRoute, an open-source intelligent routing and decision-support platform for electric vehicles. The proposed system integrates real charging-station data from Open Charge Map, road-network information from Valhalla, and GPS services provided by Traccar. Its routing engine employs an A*-based state-space search that jointly considers driving and charging times while enforcing battery state-of-charge (SoC) constraints to automatically determine feasible charging stops. The platform was designed based on an exploratory user study that identified charging-station availability and range anxiety as the primary concerns of EV drivers. Experimental evaluation using real charging infrastructure and road-network data from Northeastern Brazil demonstrates that EVRoute successfully converts trips that would otherwise be infeasible into feasible routes through automatic charging-stop planning. In the longest evaluated scenario (685 km), the platform automatically selected three charging stations while maintaining the predefined battery reserve. The results demonstrate that EVRoute bridges the gap between electric vehicle routing research and practical trip-planning tools by combining open-source implementation, real-world data integration, and intelligent route planning, supporting more equitable access to electric mobility in emerging markets.</description>
	<pubDate>2026-10-08</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 159: EVRoute: An Open-Source Intelligent Routing and Decision-Support Platform for Electric Vehicles</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/159">doi: 10.3390/automation7050159</a></p>
	<p>Authors:
		Pedro Germano Agripino Cruz
		Joylan Nunes Maciel
		Jorge Javier Gimenez Ledesma
		Marco Roberto Cavallari
		Oswaldo Hideo Ando Junior
		</p>
	<p>Electric vehicle (EV) adoption has increased the demand for intelligent trip-planning solutions capable of addressing charging infrastructure limitations and reducing range anxiety, a socio-technical barrier that disproportionately affects regions with developing charging networks. This paper presents EVRoute, an open-source intelligent routing and decision-support platform for electric vehicles. The proposed system integrates real charging-station data from Open Charge Map, road-network information from Valhalla, and GPS services provided by Traccar. Its routing engine employs an A*-based state-space search that jointly considers driving and charging times while enforcing battery state-of-charge (SoC) constraints to automatically determine feasible charging stops. The platform was designed based on an exploratory user study that identified charging-station availability and range anxiety as the primary concerns of EV drivers. Experimental evaluation using real charging infrastructure and road-network data from Northeastern Brazil demonstrates that EVRoute successfully converts trips that would otherwise be infeasible into feasible routes through automatic charging-stop planning. In the longest evaluated scenario (685 km), the platform automatically selected three charging stations while maintaining the predefined battery reserve. The results demonstrate that EVRoute bridges the gap between electric vehicle routing research and practical trip-planning tools by combining open-source implementation, real-world data integration, and intelligent route planning, supporting more equitable access to electric mobility in emerging markets.</p>
	]]></content:encoded>

	<dc:title>EVRoute: An Open-Source Intelligent Routing and Decision-Support Platform for Electric Vehicles</dc:title>
			<dc:creator>Pedro Germano Agripino Cruz</dc:creator>
			<dc:creator>Joylan Nunes Maciel</dc:creator>
			<dc:creator>Jorge Javier Gimenez Ledesma</dc:creator>
			<dc:creator>Marco Roberto Cavallari</dc:creator>
			<dc:creator>Oswaldo Hideo Ando Junior</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050159</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-08</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-08</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>159</prism:startingPage>
		<prism:doi>10.3390/automation7050159</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/159</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/158">

	<title>Automation, Vol. 7, Pages 158: Data-Driven State-Space Identification with Correlated Fuzzy Parametric Uncertainty for Prediction and Integral Control</title>
	<link>https://www.mdpi.com/2673-4052/7/5/158</link>
	<description>Data-driven models for embedded control must account for acquisition filtering and experimental uncertainty while retaining dependence among identified parameters. This work develops a framework connecting identification in filtered-state coordinates, joint fuzzy uncertainty propagation, and fixed-gain stability assessment. A structured affine discrete-time model is refined through finite-horizon prediction error, and complete bootstrap deviations define a joint radial parameter family. Set inclusion ensures nested prediction envelopes, while endpoint quadratic inequalities certify stability over continuous parameter segments. Evaluation uses 30 identification runs and a main campaign of 60 physical closed-loop trials on a DC motor (30 per controller), complemented by two further campaigns of ten trials per controller: one with an inserted series resistor and one with a sinusoidal reference. Separate calibration and testing quantify predictive coverage and interval width, while marginal-preserving permutation demonstrates the influence of parameter pairing on prediction bands and sampled stability. All 435 retained training families receive continuous stability certificates. Quantitative comparisons with N4SID, SUBNET, and a TSK fuzzy predictor identify complementary predictive strengths, and nominal versus guaranteed-cost control evaluations characterize the trade-off between tracking accuracy and voltage effort. The implemented nominal integral LQR achieves a tracking RMSE of 0.7969rad. The contribution links empirical joint uncertainty to nested propagation and verifiable stability properties within the retained model families, supported by physical control experiments. The certificates apply to the modeled linear dynamics; experimental validation remains limited to the tested platform and operating conditions.</description>
	<pubDate>2026-10-02</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 158: Data-Driven State-Space Identification with Correlated Fuzzy Parametric Uncertainty for Prediction and Integral Control</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/158">doi: 10.3390/automation7050158</a></p>
	<p>Authors:
		José R. García-Martínez
		Irving Mora-González
		Omar A. Barra-Vázquez
		Tomás Ramos-Hernández
		Rafael Rojas-Galván
		Juvenal Rodríguez-Reséndiz
		</p>
	<p>Data-driven models for embedded control must account for acquisition filtering and experimental uncertainty while retaining dependence among identified parameters. This work develops a framework connecting identification in filtered-state coordinates, joint fuzzy uncertainty propagation, and fixed-gain stability assessment. A structured affine discrete-time model is refined through finite-horizon prediction error, and complete bootstrap deviations define a joint radial parameter family. Set inclusion ensures nested prediction envelopes, while endpoint quadratic inequalities certify stability over continuous parameter segments. Evaluation uses 30 identification runs and a main campaign of 60 physical closed-loop trials on a DC motor (30 per controller), complemented by two further campaigns of ten trials per controller: one with an inserted series resistor and one with a sinusoidal reference. Separate calibration and testing quantify predictive coverage and interval width, while marginal-preserving permutation demonstrates the influence of parameter pairing on prediction bands and sampled stability. All 435 retained training families receive continuous stability certificates. Quantitative comparisons with N4SID, SUBNET, and a TSK fuzzy predictor identify complementary predictive strengths, and nominal versus guaranteed-cost control evaluations characterize the trade-off between tracking accuracy and voltage effort. The implemented nominal integral LQR achieves a tracking RMSE of 0.7969rad. The contribution links empirical joint uncertainty to nested propagation and verifiable stability properties within the retained model families, supported by physical control experiments. The certificates apply to the modeled linear dynamics; experimental validation remains limited to the tested platform and operating conditions.</p>
	]]></content:encoded>

	<dc:title>Data-Driven State-Space Identification with Correlated Fuzzy Parametric Uncertainty for Prediction and Integral Control</dc:title>
			<dc:creator>José R. García-Martínez</dc:creator>
			<dc:creator>Irving Mora-González</dc:creator>
			<dc:creator>Omar A. Barra-Vázquez</dc:creator>
			<dc:creator>Tomás Ramos-Hernández</dc:creator>
			<dc:creator>Rafael Rojas-Galván</dc:creator>
			<dc:creator>Juvenal Rodríguez-Reséndiz</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050158</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-02</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-02</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>158</prism:startingPage>
		<prism:doi>10.3390/automation7050158</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/158</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/157">

	<title>Automation, Vol. 7, Pages 157: Hybrid Thermoelectric Generation Characterization Platform</title>
	<link>https://www.mdpi.com/2673-4052/7/5/157</link>
	<description>The growing adoption of thermoelectric generators (TEGs) and solar thermoelectric generators (STEGs) for energy harvesting has increased the demand for flexible experimental platforms capable of controlled electrothermal testing under diverse thermal excitation conditions. This paper presents the development and experimental demonstration of a hybrid thermoelectric experimental platform for controlled electrothermal investigations, integrating ceramic positive temperature coefficient (PTC) heating, controlled halogen-based optical irradiation, multistage thermoelectric cooling with active liquid heat rejection, embedded PID-based thermal regulation, and synchronized thermal and electrical data acquisition. An ESP32-WROOM-32-based supervisory system provides autonomous control, remote monitoring, and the independent regulation of the heating and cooling subsystems. Under the investigated conditions, cold-side temperatures approached &amp;amp;minus;40 &amp;amp;deg;C, while hot-side temperatures approached 200 &amp;amp;deg;C under resistive excitation and approximately 170 &amp;amp;deg;C under maximum optical excitation; sustained controlled optical operation was limited to approximately 150 &amp;amp;deg;C. Four independently calibrated NTC temperature channels exhibited a maximum observed calibration deviation within &amp;amp;plusmn;1% over the evaluated ranges, representing calibration performance rather than the combined uncertainty of the complete measurement system. Timestamp-based alignment of thermal and electrical measurements resulted in a residual digital synchronization skew below 50 ms, characterizing the temporal alignment of the recorded data streams rather than the physical response time of the temperature sensors. In a transient TEG experiment, the platform produced a maximum externally measured temperature difference of approximately 161.74 K and a transient open-circuit voltage peak of 1.7818 V, demonstrating temporally aligned observation of thermal and electrical responses during controlled thermal conditioning and passive relaxation. These quantities characterize the transient system-level response under the imposed experimental boundary conditions rather than a direct determination of intrinsic TEG properties. Overall, the results demonstrate the coordinated operation of hybrid thermal excitation, multistage cooling, embedded control, and synchronized electrothermal acquisition within a single reconfigurable experimental infrastructure for TEG- and STEG-oriented investigations.</description>
	<pubDate>2026-10-02</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 157: Hybrid Thermoelectric Generation Characterization Platform</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/157">doi: 10.3390/automation7050157</a></p>
	<p>Authors:
		Eder Andrade da Silva
		Jorge Javier Gimenez Ledesma
		Marco Roberto Cavallari
		Oswaldo Hideo Ando Junior
		</p>
	<p>The growing adoption of thermoelectric generators (TEGs) and solar thermoelectric generators (STEGs) for energy harvesting has increased the demand for flexible experimental platforms capable of controlled electrothermal testing under diverse thermal excitation conditions. This paper presents the development and experimental demonstration of a hybrid thermoelectric experimental platform for controlled electrothermal investigations, integrating ceramic positive temperature coefficient (PTC) heating, controlled halogen-based optical irradiation, multistage thermoelectric cooling with active liquid heat rejection, embedded PID-based thermal regulation, and synchronized thermal and electrical data acquisition. An ESP32-WROOM-32-based supervisory system provides autonomous control, remote monitoring, and the independent regulation of the heating and cooling subsystems. Under the investigated conditions, cold-side temperatures approached &amp;amp;minus;40 &amp;amp;deg;C, while hot-side temperatures approached 200 &amp;amp;deg;C under resistive excitation and approximately 170 &amp;amp;deg;C under maximum optical excitation; sustained controlled optical operation was limited to approximately 150 &amp;amp;deg;C. Four independently calibrated NTC temperature channels exhibited a maximum observed calibration deviation within &amp;amp;plusmn;1% over the evaluated ranges, representing calibration performance rather than the combined uncertainty of the complete measurement system. Timestamp-based alignment of thermal and electrical measurements resulted in a residual digital synchronization skew below 50 ms, characterizing the temporal alignment of the recorded data streams rather than the physical response time of the temperature sensors. In a transient TEG experiment, the platform produced a maximum externally measured temperature difference of approximately 161.74 K and a transient open-circuit voltage peak of 1.7818 V, demonstrating temporally aligned observation of thermal and electrical responses during controlled thermal conditioning and passive relaxation. These quantities characterize the transient system-level response under the imposed experimental boundary conditions rather than a direct determination of intrinsic TEG properties. Overall, the results demonstrate the coordinated operation of hybrid thermal excitation, multistage cooling, embedded control, and synchronized electrothermal acquisition within a single reconfigurable experimental infrastructure for TEG- and STEG-oriented investigations.</p>
	]]></content:encoded>

	<dc:title>Hybrid Thermoelectric Generation Characterization Platform</dc:title>
			<dc:creator>Eder Andrade da Silva</dc:creator>
			<dc:creator>Jorge Javier Gimenez Ledesma</dc:creator>
			<dc:creator>Marco Roberto Cavallari</dc:creator>
			<dc:creator>Oswaldo Hideo Ando Junior</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050157</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-02</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-02</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>157</prism:startingPage>
		<prism:doi>10.3390/automation7050157</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/157</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/156">

	<title>Automation, Vol. 7, Pages 156: Recent Advances in Automated Single-Cell Manipulation Platforms: Optical Tweezers, Microfluidics, Microrobots, and Micropipette/Microneedle Systems</title>
	<link>https://www.mdpi.com/2673-4052/7/5/156</link>
	<description>Automated micromanipulation enables precise, repeatable interaction with single cells and subcellular structures. This review examines four platform archetypes defined by their dominant interaction mechanism: field-mediated optical tweezers, flow-mediated automated microfluidics, mobile microrobots, and fixed-probe micropipette/microneedle systems. Rather than treating them as equivalent robot classes, the review compares their working principles, sensing and control architectures, application fit, quantitative performance evidence, and biological trade-offs. Recent advances in closed-loop operation, intelligent perception, multimodal sensing, and hybrid workflows are analyzed, and a standardized set of reporting metrics is proposed to support objective comparison and translation.</description>
	<pubDate>2026-10-01</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 156: Recent Advances in Automated Single-Cell Manipulation Platforms: Optical Tweezers, Microfluidics, Microrobots, and Micropipette/Microneedle Systems</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/156">doi: 10.3390/automation7050156</a></p>
	<p>Authors:
		Adnan Shakoor
		</p>
	<p>Automated micromanipulation enables precise, repeatable interaction with single cells and subcellular structures. This review examines four platform archetypes defined by their dominant interaction mechanism: field-mediated optical tweezers, flow-mediated automated microfluidics, mobile microrobots, and fixed-probe micropipette/microneedle systems. Rather than treating them as equivalent robot classes, the review compares their working principles, sensing and control architectures, application fit, quantitative performance evidence, and biological trade-offs. Recent advances in closed-loop operation, intelligent perception, multimodal sensing, and hybrid workflows are analyzed, and a standardized set of reporting metrics is proposed to support objective comparison and translation.</p>
	]]></content:encoded>

	<dc:title>Recent Advances in Automated Single-Cell Manipulation Platforms: Optical Tweezers, Microfluidics, Microrobots, and Micropipette/Microneedle Systems</dc:title>
			<dc:creator>Adnan Shakoor</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050156</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-10-01</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-10-01</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Review</prism:section>
	<prism:startingPage>156</prism:startingPage>
		<prism:doi>10.3390/automation7050156</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/156</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/155">

	<title>Automation, Vol. 7, Pages 155: Automatic Road-Crack Detection with Self-Supervised YOLOv7 in Drone Imagery</title>
	<link>https://www.mdpi.com/2673-4052/7/5/155</link>
	<description>Supervised object detection models rely on large training datasets to achieve good performance. However, data labeling is often an expensive and time-consuming task. To mitigate this challenge, self-supervised methods have been utilized to learn representations from unlabeled data. Accordingly, this study uses self-supervised learning to improve the YOLOv7 model. YOLOv7 was chosen because of its reliable design and lightweight E-ELAN backbone network, which helps it learn faster without influencing the gradient path. As such, YOLOv7 undergoes two separate training stages: pre-training and fine-tuning. During pre-training, new synthetic images are automatically generated by fusing different kinds of foreground road-damage objects with various background images. In addition, a contrastive loss function was incorporated into YOLOv7 to make class-specific instances cluster together. In the second stage, the pre-trained YOLOv7 model is fine-tuned using real training images. To validate the effectiveness of the proposed approach, it has been evaluated on drone images for road-damage detection from the UAPD dataset with six classes: transverse, longitudinal, oblique, pothole, alligator, and crack repair. The numerical findings showed that self-supervised learning improved YOLOv7&amp;amp;rsquo;s performance by over eight percentage points (mAP 81.7% vs. 73.5%). The proposed approach also outperformed other detectors found in the literature, including Faster R-CNN (48.8% mAP), self-supervised DETR (20.10% mAP), YOLOv3 (68.75% mAP), and YOLOv4 (56.6% mAP).</description>
	<pubDate>2026-09-28</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 155: Automatic Road-Crack Detection with Self-Supervised YOLOv7 in Drone Imagery</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/155">doi: 10.3390/automation7050155</a></p>
	<p>Authors:
		Hussein Samma
		</p>
	<p>Supervised object detection models rely on large training datasets to achieve good performance. However, data labeling is often an expensive and time-consuming task. To mitigate this challenge, self-supervised methods have been utilized to learn representations from unlabeled data. Accordingly, this study uses self-supervised learning to improve the YOLOv7 model. YOLOv7 was chosen because of its reliable design and lightweight E-ELAN backbone network, which helps it learn faster without influencing the gradient path. As such, YOLOv7 undergoes two separate training stages: pre-training and fine-tuning. During pre-training, new synthetic images are automatically generated by fusing different kinds of foreground road-damage objects with various background images. In addition, a contrastive loss function was incorporated into YOLOv7 to make class-specific instances cluster together. In the second stage, the pre-trained YOLOv7 model is fine-tuned using real training images. To validate the effectiveness of the proposed approach, it has been evaluated on drone images for road-damage detection from the UAPD dataset with six classes: transverse, longitudinal, oblique, pothole, alligator, and crack repair. The numerical findings showed that self-supervised learning improved YOLOv7&amp;amp;rsquo;s performance by over eight percentage points (mAP 81.7% vs. 73.5%). The proposed approach also outperformed other detectors found in the literature, including Faster R-CNN (48.8% mAP), self-supervised DETR (20.10% mAP), YOLOv3 (68.75% mAP), and YOLOv4 (56.6% mAP).</p>
	]]></content:encoded>

	<dc:title>Automatic Road-Crack Detection with Self-Supervised YOLOv7 in Drone Imagery</dc:title>
			<dc:creator>Hussein Samma</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050155</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-28</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-28</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>155</prism:startingPage>
		<prism:doi>10.3390/automation7050155</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/155</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/154">

	<title>Automation, Vol. 7, Pages 154: Quasi-LPV Control of Multiple Quadrotors for Cooperative Suspended Payload Transportation</title>
	<link>https://www.mdpi.com/2673-4052/7/5/154</link>
	<description>This paper presents a quasi-linear parameter-varying (quasi-LPV) control framework for the cooperative transportation of a cable-suspended payload using multiple quadrotors. The main objective is to exploit systematic linear-system control techniques to reduce the complexity of the controller design while retaining the relevant nonlinear dependencies of the multi-vehicle system. The proposed architecture is organized into two interconnected levels: a payload-level force allocation stage and a vehicle-level quasi-LPV control stage. At the payload level, the required resultant force is distributed among the suspension cables through a regularized least-squares formulation. At the vehicle level, the nonlinear quadrotor dynamics are represented by a polytopic quasi-LPV model using measurable scheduling variables. An integral state-feedback controller is synthesized through linear matrix inequalities (LMIs), incorporating an H&amp;amp;infin; criterion to attenuate the effect of cable-induced interaction forces on the tracking performance. Numerical simulations are performed for formations composed of three, five, and ten quadrotors following circular, figure-eight, and three-dimensional reference trajectories, respectively. The results show accurate payload and vehicle trajectory tracking, bounded cable-force allocation errors, and consistent performance as the number of cooperating vehicles increases. These results demonstrate that the quasi-LPV formulation enables linear control-design tools to be effectively applied to a nonlinear and strongly coupled cooperative aerial transportation system.</description>
	<pubDate>2026-09-25</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 154: Quasi-LPV Control of Multiple Quadrotors for Cooperative Suspended Payload Transportation</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/154">doi: 10.3390/automation7050154</a></p>
	<p>Authors:
		Gerardo Ortiz-Torres
		Manuel A. Zurita-Gil
		Jesse Y. Rumbo-Morales
		Felipe D. J. Sorcia-Vázquez
		Alan Francisco Pérez-Vidal
		Jorge Aurelio Brizuela-Mendoza
		Jorge A. Adriano-Colesio
		</p>
	<p>This paper presents a quasi-linear parameter-varying (quasi-LPV) control framework for the cooperative transportation of a cable-suspended payload using multiple quadrotors. The main objective is to exploit systematic linear-system control techniques to reduce the complexity of the controller design while retaining the relevant nonlinear dependencies of the multi-vehicle system. The proposed architecture is organized into two interconnected levels: a payload-level force allocation stage and a vehicle-level quasi-LPV control stage. At the payload level, the required resultant force is distributed among the suspension cables through a regularized least-squares formulation. At the vehicle level, the nonlinear quadrotor dynamics are represented by a polytopic quasi-LPV model using measurable scheduling variables. An integral state-feedback controller is synthesized through linear matrix inequalities (LMIs), incorporating an H&amp;amp;infin; criterion to attenuate the effect of cable-induced interaction forces on the tracking performance. Numerical simulations are performed for formations composed of three, five, and ten quadrotors following circular, figure-eight, and three-dimensional reference trajectories, respectively. The results show accurate payload and vehicle trajectory tracking, bounded cable-force allocation errors, and consistent performance as the number of cooperating vehicles increases. These results demonstrate that the quasi-LPV formulation enables linear control-design tools to be effectively applied to a nonlinear and strongly coupled cooperative aerial transportation system.</p>
	]]></content:encoded>

	<dc:title>Quasi-LPV Control of Multiple Quadrotors for Cooperative Suspended Payload Transportation</dc:title>
			<dc:creator>Gerardo Ortiz-Torres</dc:creator>
			<dc:creator>Manuel A. Zurita-Gil</dc:creator>
			<dc:creator>Jesse Y. Rumbo-Morales</dc:creator>
			<dc:creator>Felipe D. J. Sorcia-Vázquez</dc:creator>
			<dc:creator>Alan Francisco Pérez-Vidal</dc:creator>
			<dc:creator>Jorge Aurelio Brizuela-Mendoza</dc:creator>
			<dc:creator>Jorge A. Adriano-Colesio</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050154</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-25</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-25</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>154</prism:startingPage>
		<prism:doi>10.3390/automation7050154</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/154</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/153">

	<title>Automation, Vol. 7, Pages 153: WQ-ERT: Water Quality Prediction in Low-Resource Domains Using Gaussian Noise-Augmented Transformer Encoder: Case Study in Dili, Timor-Leste</title>
	<link>https://www.mdpi.com/2673-4052/7/5/153</link>
	<description>Evaluating water quality is fundamental to ensuring that freshwater ecosystems support the well-being of all life on Earth. The present study uses physicochemical parameters to assess groundwater in Dili, Timor-Leste, from 2011 to 2018. Water quality was determined using the conventional Water Quality Index (WQI), aggregated via a weighted sum function aligned with the World Health Organization (WHO). The study aims to predict the WQI and water quality status, expressed as Water Quality Classification (WQC), to support sustainable water resource management. In low-resource domains, it is crucial to address challenges related to small dataset size and class imbalance in the sample observations. However, deep learning models are highly dependent on the availability and quality of training data. This paper builds upon prior data augmentation strategies for water quality data analysis, aiming to overcome the constraints of a small dataset in predicting the WQI and WQC. To address these issues, the study extends prior data augmentation strategies and introduces WQ-ERT (Water Quality Encoder Representations from Transformer). This Transformer-based model applies Gaussian noise augmentation and a class-balancing strategy. The model has been pre-trained on a large Indian water quality dataset with similar soil-type characteristics and fine-tuned on the Timor-Leste water quality dataset, using a Gaussian noise value of &amp;amp;alpha; = 0.20 for the Indian datasets, whereas for Timor-Leste, it uses a lower Gaussian noise value of &amp;amp;alpha; = 0.05. Furthermore, to address label imbalance in the classification target, a random search-based class-balancing technique is employed during fine-tuning. We conducted six experiments to evaluate our proposed model and applied five-fold cross-validation. The results demonstrate that noise augmentation substantially improves WQC performance, achieving 97.0% balanced accuracy, and WQI prediction reaches an R2 of 0.703, indicating better generalization across the dataset. Class-balancing techniques enhance WQC, particularly in balanced accuracy, F1-score, and the Matthews Correlation Coefficient (MCC). This configuration reduces false negatives by 86.05% for water quality classification, though excessive weighting slightly lowers precision. In WQI prediction, the experiment shows a 7.92% reduction in MAE, improving few-shot adaptation to Timor-Leste without distortion.</description>
	<pubDate>2026-09-24</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 153: WQ-ERT: Water Quality Prediction in Low-Resource Domains Using Gaussian Noise-Augmented Transformer Encoder: Case Study in Dili, Timor-Leste</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/153">doi: 10.3390/automation7050153</a></p>
	<p>Authors:
		Zulmira Ximenes da Costa
		Daito Takahashi
		Yuichi Nishida
		Yuya Murai
		Floris Cornelis Boogaard
		Satoshi Tamura
		</p>
	<p>Evaluating water quality is fundamental to ensuring that freshwater ecosystems support the well-being of all life on Earth. The present study uses physicochemical parameters to assess groundwater in Dili, Timor-Leste, from 2011 to 2018. Water quality was determined using the conventional Water Quality Index (WQI), aggregated via a weighted sum function aligned with the World Health Organization (WHO). The study aims to predict the WQI and water quality status, expressed as Water Quality Classification (WQC), to support sustainable water resource management. In low-resource domains, it is crucial to address challenges related to small dataset size and class imbalance in the sample observations. However, deep learning models are highly dependent on the availability and quality of training data. This paper builds upon prior data augmentation strategies for water quality data analysis, aiming to overcome the constraints of a small dataset in predicting the WQI and WQC. To address these issues, the study extends prior data augmentation strategies and introduces WQ-ERT (Water Quality Encoder Representations from Transformer). This Transformer-based model applies Gaussian noise augmentation and a class-balancing strategy. The model has been pre-trained on a large Indian water quality dataset with similar soil-type characteristics and fine-tuned on the Timor-Leste water quality dataset, using a Gaussian noise value of &amp;amp;alpha; = 0.20 for the Indian datasets, whereas for Timor-Leste, it uses a lower Gaussian noise value of &amp;amp;alpha; = 0.05. Furthermore, to address label imbalance in the classification target, a random search-based class-balancing technique is employed during fine-tuning. We conducted six experiments to evaluate our proposed model and applied five-fold cross-validation. The results demonstrate that noise augmentation substantially improves WQC performance, achieving 97.0% balanced accuracy, and WQI prediction reaches an R2 of 0.703, indicating better generalization across the dataset. Class-balancing techniques enhance WQC, particularly in balanced accuracy, F1-score, and the Matthews Correlation Coefficient (MCC). This configuration reduces false negatives by 86.05% for water quality classification, though excessive weighting slightly lowers precision. In WQI prediction, the experiment shows a 7.92% reduction in MAE, improving few-shot adaptation to Timor-Leste without distortion.</p>
	]]></content:encoded>

	<dc:title>WQ-ERT: Water Quality Prediction in Low-Resource Domains Using Gaussian Noise-Augmented Transformer Encoder: Case Study in Dili, Timor-Leste</dc:title>
			<dc:creator>Zulmira Ximenes da Costa</dc:creator>
			<dc:creator>Daito Takahashi</dc:creator>
			<dc:creator>Yuichi Nishida</dc:creator>
			<dc:creator>Yuya Murai</dc:creator>
			<dc:creator>Floris Cornelis Boogaard</dc:creator>
			<dc:creator>Satoshi Tamura</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050153</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-24</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-24</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>153</prism:startingPage>
		<prism:doi>10.3390/automation7050153</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/153</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/152">

	<title>Automation, Vol. 7, Pages 152: An Evidence-Centered Decision-Making Framework for Explainable Autonomous Recovery in Self-Healing Software Systems</title>
	<link>https://www.mdpi.com/2673-4052/7/5/152</link>
	<description>Self-healing software systems have become an important approach for improving the reliability and availability of modern distributed applications. However, existing autonomous recovery methods are predominantly based on predefined rules or data-driven models, which often provide limited transparency, weak evidence traceability, and insufficient justification of recovery decisions. This paper proposes an Evidence-Centered Decision-Making (ECDM) framework for explainable autonomous recovery in self-healing software systems. The proposed framework transforms runtime observations into structured evidence, constructs an Evidence Graph, generates recovery hypotheses, and evaluates alternative recovery actions using the proposed Hierarchical Decision Confidence Model (HDCM). The framework was evaluated through an experimental campaign consisting of 110 recovery-decision episodes across five representative failure scenarios and multiple fault severity levels. In the confirmatory evaluation of 110 recovery-decision episodes, ECDM achieved an overall decision accuracy of 85.45%, with the highest scenario-level accuracy observed for CPU Resource Exhaustion and Service Dependency Failure. The Safety Gate admitted 49 decisions for autonomous handling, including 34 active recovery actions. All 34 executed active recovery actions satisfied the predefined recovery-verification criteria, with a median Recovery Verification Latency of 1048.9 ms, while autonomous coverage varied substantially across scenarios. The proposed ECDM framework provides a traceable recovery-decision framework by combining evidence modeling, hierarchical confidence estimation, and structured decision justification.</description>
	<pubDate>2026-09-22</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 152: An Evidence-Centered Decision-Making Framework for Explainable Autonomous Recovery in Self-Healing Software Systems</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/152">doi: 10.3390/automation7050152</a></p>
	<p>Authors:
		Oleh Pitsun
		</p>
	<p>Self-healing software systems have become an important approach for improving the reliability and availability of modern distributed applications. However, existing autonomous recovery methods are predominantly based on predefined rules or data-driven models, which often provide limited transparency, weak evidence traceability, and insufficient justification of recovery decisions. This paper proposes an Evidence-Centered Decision-Making (ECDM) framework for explainable autonomous recovery in self-healing software systems. The proposed framework transforms runtime observations into structured evidence, constructs an Evidence Graph, generates recovery hypotheses, and evaluates alternative recovery actions using the proposed Hierarchical Decision Confidence Model (HDCM). The framework was evaluated through an experimental campaign consisting of 110 recovery-decision episodes across five representative failure scenarios and multiple fault severity levels. In the confirmatory evaluation of 110 recovery-decision episodes, ECDM achieved an overall decision accuracy of 85.45%, with the highest scenario-level accuracy observed for CPU Resource Exhaustion and Service Dependency Failure. The Safety Gate admitted 49 decisions for autonomous handling, including 34 active recovery actions. All 34 executed active recovery actions satisfied the predefined recovery-verification criteria, with a median Recovery Verification Latency of 1048.9 ms, while autonomous coverage varied substantially across scenarios. The proposed ECDM framework provides a traceable recovery-decision framework by combining evidence modeling, hierarchical confidence estimation, and structured decision justification.</p>
	]]></content:encoded>

	<dc:title>An Evidence-Centered Decision-Making Framework for Explainable Autonomous Recovery in Self-Healing Software Systems</dc:title>
			<dc:creator>Oleh Pitsun</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050152</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-22</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-22</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>152</prism:startingPage>
		<prism:doi>10.3390/automation7050152</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/152</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/151">

	<title>Automation, Vol. 7, Pages 151: A Hybrid Methodology for Intelligent Decision Support in Catalytic Cracking Under Uncertainty</title>
	<link>https://www.mdpi.com/2673-4052/7/5/151</link>
	<description>This article presents a hybrid methodology for the intelligent control of the reactor-regenerator unit of a residual fluid catalytic cracking (RFCC) plant under conditions of uncertain input information. The main challenge in catalytic cracking process control is the instability of feedstock composition and the delay in laboratory data on product quality, which complicates the operational management of the RFCC process while maintaining the required gasoline quality. The objective of this study is to support effective management of the gasoline production process by maximizing the yield of high-quality gasoline while keeping its density within specified limits. The proposed architecture combines a regression model built on six dominant parameters selected through Pearson&amp;amp;rsquo;s correlation analysis, a Mamdani-type fuzzy inference system with a database of 26 expert rules, and a machine-learning module (Random Forest and gradient boosting) that corrects the residual error of the combined regression-fuzzy model, allowing the control system to adapt to changing process conditions. Testing the developed models on real data from the Shymkent Oil Refinery shows that the hybrid model reduces the root-mean-square error from 0.1973 (regression model alone) to 0.0161 and the mean absolute percentage error from 0.328% to 0.028%, while the coefficient of determination increases from 0.9895 to 0.9999; this improved accuracy is estimated to correspond to an approximate 3.2% increase in achievable gasoline yield and a 0.4% improvement in density stability. The developed decision support system for catalytic cracking process control, based on the proposed methodology, acts as a &amp;amp;ldquo;virtual analyzer&amp;amp;rdquo;, ensuring effective control in real time without the use of expensive in-line quality control devices.</description>
	<pubDate>2026-09-21</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 151: A Hybrid Methodology for Intelligent Decision Support in Catalytic Cracking Under Uncertainty</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/151">doi: 10.3390/automation7050151</a></p>
	<p>Authors:
		Narkez Boranbayeva
		Batyr Orazbayev
		Madyar Kabibullin
		Togzhan Kenzhebayeva
		Gulnara Abitova
		Aiman Kaliyeva
		Assylkhan Zhanekeshova
		</p>
	<p>This article presents a hybrid methodology for the intelligent control of the reactor-regenerator unit of a residual fluid catalytic cracking (RFCC) plant under conditions of uncertain input information. The main challenge in catalytic cracking process control is the instability of feedstock composition and the delay in laboratory data on product quality, which complicates the operational management of the RFCC process while maintaining the required gasoline quality. The objective of this study is to support effective management of the gasoline production process by maximizing the yield of high-quality gasoline while keeping its density within specified limits. The proposed architecture combines a regression model built on six dominant parameters selected through Pearson&amp;amp;rsquo;s correlation analysis, a Mamdani-type fuzzy inference system with a database of 26 expert rules, and a machine-learning module (Random Forest and gradient boosting) that corrects the residual error of the combined regression-fuzzy model, allowing the control system to adapt to changing process conditions. Testing the developed models on real data from the Shymkent Oil Refinery shows that the hybrid model reduces the root-mean-square error from 0.1973 (regression model alone) to 0.0161 and the mean absolute percentage error from 0.328% to 0.028%, while the coefficient of determination increases from 0.9895 to 0.9999; this improved accuracy is estimated to correspond to an approximate 3.2% increase in achievable gasoline yield and a 0.4% improvement in density stability. The developed decision support system for catalytic cracking process control, based on the proposed methodology, acts as a &amp;amp;ldquo;virtual analyzer&amp;amp;rdquo;, ensuring effective control in real time without the use of expensive in-line quality control devices.</p>
	]]></content:encoded>

	<dc:title>A Hybrid Methodology for Intelligent Decision Support in Catalytic Cracking Under Uncertainty</dc:title>
			<dc:creator>Narkez Boranbayeva</dc:creator>
			<dc:creator>Batyr Orazbayev</dc:creator>
			<dc:creator>Madyar Kabibullin</dc:creator>
			<dc:creator>Togzhan Kenzhebayeva</dc:creator>
			<dc:creator>Gulnara Abitova</dc:creator>
			<dc:creator>Aiman Kaliyeva</dc:creator>
			<dc:creator>Assylkhan Zhanekeshova</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050151</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-21</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-21</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>151</prism:startingPage>
		<prism:doi>10.3390/automation7050151</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/151</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/150">

	<title>Automation, Vol. 7, Pages 150: Symmetry-Based Decoupled Extended Reality and Digital Twin Framework for Robotic Manipulators</title>
	<link>https://www.mdpi.com/2673-4052/7/5/150</link>
	<description>This paper presents a decoupled Extended Reality (XR) and Digital Twin (DT) framework for robotic manipulators, leveraging mathematical, physical, and topological symmetries to achieve stable cyber&amp;amp;ndash;physical synchronization. Traditional robotic teleoperation architectures suffer from tight coupling between low-level control, communication, and visualization interfaces, making them vulnerable to network latency and platform dependencies. We propose an open, modular framework that structurally decouples physical and virtual counterparts. By establishing a system-theoretic formulation of Digital Twin symmetry, we represent the virtual robot as an isomorphic state-space mirror of the physical manipulator. Communication is mediated by an asynchronous, network topology using a lightweight Message Queuing Telemetry Transport (MQTT) broker, ensuring topic-based communication symmetry. This decoupled structure enables a highly flexible, many-to-many evaluation paradigm, allowing researchers to interchangeably pair a single physical robot with diverse user interfaces (UIs) or control multiple heterogeneous robots (physical and simulated) with a single virtual interface. Experimental validation on a 7-Degree-of-Freedom (7-DOF) Franka Emika Panda robot and a unified data collector confirms high-fidelity tracking, with a symmetric wired and wireless Round-Trip Times (RTT) averaging 47 ms. Finally, we discuss the challenges of transitioning this laboratory-scale architecture to high-concurrency industrial settings, addressing single-broker vulnerabilities, network jitter, and dynamic calibration drift.</description>
	<pubDate>2026-09-21</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 150: Symmetry-Based Decoupled Extended Reality and Digital Twin Framework for Robotic Manipulators</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/150">doi: 10.3390/automation7050150</a></p>
	<p>Authors:
		Daniel Stoyanov
		</p>
	<p>This paper presents a decoupled Extended Reality (XR) and Digital Twin (DT) framework for robotic manipulators, leveraging mathematical, physical, and topological symmetries to achieve stable cyber&amp;amp;ndash;physical synchronization. Traditional robotic teleoperation architectures suffer from tight coupling between low-level control, communication, and visualization interfaces, making them vulnerable to network latency and platform dependencies. We propose an open, modular framework that structurally decouples physical and virtual counterparts. By establishing a system-theoretic formulation of Digital Twin symmetry, we represent the virtual robot as an isomorphic state-space mirror of the physical manipulator. Communication is mediated by an asynchronous, network topology using a lightweight Message Queuing Telemetry Transport (MQTT) broker, ensuring topic-based communication symmetry. This decoupled structure enables a highly flexible, many-to-many evaluation paradigm, allowing researchers to interchangeably pair a single physical robot with diverse user interfaces (UIs) or control multiple heterogeneous robots (physical and simulated) with a single virtual interface. Experimental validation on a 7-Degree-of-Freedom (7-DOF) Franka Emika Panda robot and a unified data collector confirms high-fidelity tracking, with a symmetric wired and wireless Round-Trip Times (RTT) averaging 47 ms. Finally, we discuss the challenges of transitioning this laboratory-scale architecture to high-concurrency industrial settings, addressing single-broker vulnerabilities, network jitter, and dynamic calibration drift.</p>
	]]></content:encoded>

	<dc:title>Symmetry-Based Decoupled Extended Reality and Digital Twin Framework for Robotic Manipulators</dc:title>
			<dc:creator>Daniel Stoyanov</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050150</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-21</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-21</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>150</prism:startingPage>
		<prism:doi>10.3390/automation7050150</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/150</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/149">

	<title>Automation, Vol. 7, Pages 149: Transfer Learning-Based Optimization of the Ore Grinding Process Using Artificial Neural Networks</title>
	<link>https://www.mdpi.com/2673-4052/7/5/149</link>
	<description>This study proposes a novel and comprehensive mathematical modeling approach designed for the multi-criteria optimization of ore-grinding processes within industrial electric drive mills. In the mining industry, the development of high-fidelity predictive models is frequently impeded by the scarcity of high-quality experimental data&amp;amp;mdash;a limitation that traditionally undermines the reliability of process control systems. To address this challenge, this research implements a sophisticated data augmentation strategy integrated with advanced transfer learning methodologies. The proposed framework begins with the artificial synthesis of datasets to enrich the information base, which is subsequently utilized to pre-train, evaluate, and refine various deep neural network architectures. By systematically identifying the most effective architecture for this specific technological context, the study fine-tunes the selected models using domain-specific parameters to ensure maximum operational relevance. The resulting robust neural network framework successfully solves complex optimization problems related to energy efficiency and throughput maximization. Quantitative results demonstrate that the proposed transfer learning paradigm accurately captures the non-linear dynamics of the grinding process, providing a physics-informed surrogate neural control model that serves as a cost-effective alternative to expensive and logistically demanding physical experiments. Ultimately, this research offers a scalable and computationally efficient framework for enhancing the operational performance and sustainability of raw ore-grinding systems in modern mineral processing plants.</description>
	<pubDate>2026-09-20</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 149: Transfer Learning-Based Optimization of the Ore Grinding Process Using Artificial Neural Networks</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/149">doi: 10.3390/automation7050149</a></p>
	<p>Authors:
		Marinka Baghdasaryan
		Armenak Babayan
		Davit Melkonyan
		</p>
	<p>This study proposes a novel and comprehensive mathematical modeling approach designed for the multi-criteria optimization of ore-grinding processes within industrial electric drive mills. In the mining industry, the development of high-fidelity predictive models is frequently impeded by the scarcity of high-quality experimental data&amp;amp;mdash;a limitation that traditionally undermines the reliability of process control systems. To address this challenge, this research implements a sophisticated data augmentation strategy integrated with advanced transfer learning methodologies. The proposed framework begins with the artificial synthesis of datasets to enrich the information base, which is subsequently utilized to pre-train, evaluate, and refine various deep neural network architectures. By systematically identifying the most effective architecture for this specific technological context, the study fine-tunes the selected models using domain-specific parameters to ensure maximum operational relevance. The resulting robust neural network framework successfully solves complex optimization problems related to energy efficiency and throughput maximization. Quantitative results demonstrate that the proposed transfer learning paradigm accurately captures the non-linear dynamics of the grinding process, providing a physics-informed surrogate neural control model that serves as a cost-effective alternative to expensive and logistically demanding physical experiments. Ultimately, this research offers a scalable and computationally efficient framework for enhancing the operational performance and sustainability of raw ore-grinding systems in modern mineral processing plants.</p>
	]]></content:encoded>

	<dc:title>Transfer Learning-Based Optimization of the Ore Grinding Process Using Artificial Neural Networks</dc:title>
			<dc:creator>Marinka Baghdasaryan</dc:creator>
			<dc:creator>Armenak Babayan</dc:creator>
			<dc:creator>Davit Melkonyan</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050149</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-20</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-20</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>149</prism:startingPage>
		<prism:doi>10.3390/automation7050149</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/149</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/148">

	<title>Automation, Vol. 7, Pages 148: Eliminating Cold Start in Smart Home Lighting Control via Digital Twin-Trained Q-Learning</title>
	<link>https://www.mdpi.com/2673-4052/7/5/148</link>
	<description>Smart home lighting systems are commonly based on static rule-based automations that cannot be effectively adjusted to changing daylight conditions and room-specific optical characteristics. Although reinforcement learning can provide adaptive control, learning directly in an occupied home can require disruptive exploratory actions during the cold-start period. This study aims to develop a lightweight cloud-independent controller that provides adaptive illuminance regulation without online reinforcement learning exploration. A room-specific empirical simulation environment based on preprocessed natural illuminance measurements was used to train a tabular Q-learning policy offline for 2500 episodes. The controller uses a compact 5&amp;amp;times;2&amp;amp;times;7 Q-table that maps discretized illuminance error and occupancy to relative brightness adjustments. During deployment, the learned policy is used only when occupancy is detected, whereas a deterministic 30-percentage-point brightness ramp-down rule is applied in the unoccupied state. The controller was implemented in Node-RED within Home Assistant on a Raspberry Pi 5 and connected to a BH1750 illuminance sensor and a Philips Hue bulb. In the evaluated simulated under-illuminated scenarios, the learned policy reached the 25 &amp;amp;plusmn; 2 lx target band within at most five control steps and avoided the initial exploratory behavior observed in Q-learning trained from scratch. Although the proportional rule-based baseline achieved lower MAE and RMSE in the evaluated simulations, the proposed method provides an interpretable, locally executable policy that eliminates online cold-start exploration and can be retrained for a specific lighting environment.</description>
	<pubDate>2026-09-20</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 148: Eliminating Cold Start in Smart Home Lighting Control via Digital Twin-Trained Q-Learning</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/148">doi: 10.3390/automation7050148</a></p>
	<p>Authors:
		Tomáš Sladčík
		Hashim Habiballa
		</p>
	<p>Smart home lighting systems are commonly based on static rule-based automations that cannot be effectively adjusted to changing daylight conditions and room-specific optical characteristics. Although reinforcement learning can provide adaptive control, learning directly in an occupied home can require disruptive exploratory actions during the cold-start period. This study aims to develop a lightweight cloud-independent controller that provides adaptive illuminance regulation without online reinforcement learning exploration. A room-specific empirical simulation environment based on preprocessed natural illuminance measurements was used to train a tabular Q-learning policy offline for 2500 episodes. The controller uses a compact 5&amp;amp;times;2&amp;amp;times;7 Q-table that maps discretized illuminance error and occupancy to relative brightness adjustments. During deployment, the learned policy is used only when occupancy is detected, whereas a deterministic 30-percentage-point brightness ramp-down rule is applied in the unoccupied state. The controller was implemented in Node-RED within Home Assistant on a Raspberry Pi 5 and connected to a BH1750 illuminance sensor and a Philips Hue bulb. In the evaluated simulated under-illuminated scenarios, the learned policy reached the 25 &amp;amp;plusmn; 2 lx target band within at most five control steps and avoided the initial exploratory behavior observed in Q-learning trained from scratch. Although the proportional rule-based baseline achieved lower MAE and RMSE in the evaluated simulations, the proposed method provides an interpretable, locally executable policy that eliminates online cold-start exploration and can be retrained for a specific lighting environment.</p>
	]]></content:encoded>

	<dc:title>Eliminating Cold Start in Smart Home Lighting Control via Digital Twin-Trained Q-Learning</dc:title>
			<dc:creator>Tomáš Sladčík</dc:creator>
			<dc:creator>Hashim Habiballa</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050148</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-20</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-20</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>148</prism:startingPage>
		<prism:doi>10.3390/automation7050148</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/148</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/147">

	<title>Automation, Vol. 7, Pages 147: A Lightweight RGB-Thermal Framework Toward UAV-Assisted Day&amp;ndash;Night Cement Crack Inspection Using MobileNetV4-CBAM and Enhanced SIFT Localization</title>
	<link>https://www.mdpi.com/2673-4052/7/5/147</link>
	<description>Reliable cement crack inspection is essential for structural health monitoring, particularly under challenging illumination conditions such as nighttime and low-light environments. Existing detection-based methods often require extensive bounding-box annotations, high computational resources, and show limited robustness in poor visibility conditions. This paper presents a lightweight RGB-thermal framework that integrates a MobileNetV4 classifier with a Convolutional Block Attention Module (CBAM) and an Enhanced Scale-Invariant Feature Transform (SIFT) localization algorithm for day&amp;amp;ndash;night crack inspection. Unlike conventional object detection approaches, the proposed method enables crack classification and localization without manual bounding-box annotation. The framework is evaluated using RGB and thermal images from three datasets, including the GYU-DET benchmark and two newly developed thermal crack datasets. On GYU-DET, the proposed classifier achieves a weighted F1-score of 83.16% using only region-of-interest-level labels, in contrast to the bounding-box-supervised YOLOv11n baseline on the same benchmark (mAP@0.5 = 0.568). Because the two approaches solve different sub-problems&amp;amp;ndash;joint detection-and-localization versus region-level classification&amp;amp;ndash;their results are reported as a capability comparison rather than a numerical improvement. Notably, crack, one of the two most difficult categories for the bounding-box detector, becomes one of the most reliably recognized categories once localization ambiguity is removed by the proposed decoupled design. For thermal morphology recognition, the original ROI-level evaluation yields weighted F1-scores of 92.25% and 92.69% on Datasets 2 and 3, respectively, within the collected crack set. A five-run crack-disjoint evaluation on Dataset 2, restricted to six morphology classes with sufficient numbers of independent physical cracks, yields an accuracy of 82.76&amp;amp;plusmn;3.43% and a macro-F1 of 82.03&amp;amp;plusmn;3.60%. End-to-end localization and classification from raw thermograms yields micro-F1 scores of 78.62% and 75.24% on Datasets 2 and 3, respectively. The Enhanced SIFT localization method, which requires no training, achieves F1-scores of 0.888 and 0.867 under passive and active thermal conditions, respectively, while operating within a 200 MB memory footprint on an entry-level 4 GB GPU. The lightweight and illumination-robust design enables reliable nighttime crack inspection and supports UAV-assisted and edge-based structural health monitoring applications.</description>
	<pubDate>2026-09-19</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 147: A Lightweight RGB-Thermal Framework Toward UAV-Assisted Day&amp;ndash;Night Cement Crack Inspection Using MobileNetV4-CBAM and Enhanced SIFT Localization</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/147">doi: 10.3390/automation7050147</a></p>
	<p>Authors:
		Ata Jahangir Moshayedi
		Ba Tu Phung
		Amir Sohail Khan
		Quoc Nguyen
		Seyed Ali Eftekhari
		Amin Kolahdooz
		David Bassir
		</p>
	<p>Reliable cement crack inspection is essential for structural health monitoring, particularly under challenging illumination conditions such as nighttime and low-light environments. Existing detection-based methods often require extensive bounding-box annotations, high computational resources, and show limited robustness in poor visibility conditions. This paper presents a lightweight RGB-thermal framework that integrates a MobileNetV4 classifier with a Convolutional Block Attention Module (CBAM) and an Enhanced Scale-Invariant Feature Transform (SIFT) localization algorithm for day&amp;amp;ndash;night crack inspection. Unlike conventional object detection approaches, the proposed method enables crack classification and localization without manual bounding-box annotation. The framework is evaluated using RGB and thermal images from three datasets, including the GYU-DET benchmark and two newly developed thermal crack datasets. On GYU-DET, the proposed classifier achieves a weighted F1-score of 83.16% using only region-of-interest-level labels, in contrast to the bounding-box-supervised YOLOv11n baseline on the same benchmark (mAP@0.5 = 0.568). Because the two approaches solve different sub-problems&amp;amp;ndash;joint detection-and-localization versus region-level classification&amp;amp;ndash;their results are reported as a capability comparison rather than a numerical improvement. Notably, crack, one of the two most difficult categories for the bounding-box detector, becomes one of the most reliably recognized categories once localization ambiguity is removed by the proposed decoupled design. For thermal morphology recognition, the original ROI-level evaluation yields weighted F1-scores of 92.25% and 92.69% on Datasets 2 and 3, respectively, within the collected crack set. A five-run crack-disjoint evaluation on Dataset 2, restricted to six morphology classes with sufficient numbers of independent physical cracks, yields an accuracy of 82.76&amp;amp;plusmn;3.43% and a macro-F1 of 82.03&amp;amp;plusmn;3.60%. End-to-end localization and classification from raw thermograms yields micro-F1 scores of 78.62% and 75.24% on Datasets 2 and 3, respectively. The Enhanced SIFT localization method, which requires no training, achieves F1-scores of 0.888 and 0.867 under passive and active thermal conditions, respectively, while operating within a 200 MB memory footprint on an entry-level 4 GB GPU. The lightweight and illumination-robust design enables reliable nighttime crack inspection and supports UAV-assisted and edge-based structural health monitoring applications.</p>
	]]></content:encoded>

	<dc:title>A Lightweight RGB-Thermal Framework Toward UAV-Assisted Day&amp;amp;ndash;Night Cement Crack Inspection Using MobileNetV4-CBAM and Enhanced SIFT Localization</dc:title>
			<dc:creator>Ata Jahangir Moshayedi</dc:creator>
			<dc:creator>Ba Tu Phung</dc:creator>
			<dc:creator>Amir Sohail Khan</dc:creator>
			<dc:creator>Quoc Nguyen</dc:creator>
			<dc:creator>Seyed Ali Eftekhari</dc:creator>
			<dc:creator>Amin Kolahdooz</dc:creator>
			<dc:creator>David Bassir</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050147</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-19</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-19</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>147</prism:startingPage>
		<prism:doi>10.3390/automation7050147</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/147</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/146">

	<title>Automation, Vol. 7, Pages 146: A Generalized Framework for CAD Design Automation of Configurable Products Through CAD&amp;ndash;PLM Integration with Quantitative Industrial Validation</title>
	<link>https://www.mdpi.com/2673-4052/7/5/146</link>
	<description>This study addresses the growing demand for efficient customization in industrial design by proposing a structured approach to automate Computer Aided Design (CAD) assembly generation and documentation. The objective is to reduce repetitive engineering effort while maintaining consistency and integration within existing digital environments. The methodology combines parametric modelling, configuration-driven design, and Visual Basic for Applications (VBA)-based scripting through CAD Application Programming Interfaces, supported by integration with a Product Lifecycle Management system. A structured data model and reorganised modular CAD architecture enable automated generation of assemblies, selection of configurations, and creation of technical drawings. The approach is validated through an industrial case study involving configurable mechanical assemblies, where performance is evaluated using a quantitative time-based framework. The results show a reduction in total configuration time from 8 to 16 h to approximately 2&amp;amp;ndash;3 h, corresponding to an average time saving of 79% and a significant decrease in effective engineering workload. Assessment of scalability indicates that engineering effort becomes largely independent of the number of configurations, while economic evaluation indicates strong return-on-investment potential under typical industrial conditions. The findings confirm that systematic CAD automation can substantially improve productivity and consistency in configuration-driven design processes, provided that parametric models and data structures are properly standardised.</description>
	<pubDate>2026-09-17</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 146: A Generalized Framework for CAD Design Automation of Configurable Products Through CAD&amp;ndash;PLM Integration with Quantitative Industrial Validation</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/146">doi: 10.3390/automation7050146</a></p>
	<p>Authors:
		Jakub Radkovsky
		Adam Becvar
		</p>
	<p>This study addresses the growing demand for efficient customization in industrial design by proposing a structured approach to automate Computer Aided Design (CAD) assembly generation and documentation. The objective is to reduce repetitive engineering effort while maintaining consistency and integration within existing digital environments. The methodology combines parametric modelling, configuration-driven design, and Visual Basic for Applications (VBA)-based scripting through CAD Application Programming Interfaces, supported by integration with a Product Lifecycle Management system. A structured data model and reorganised modular CAD architecture enable automated generation of assemblies, selection of configurations, and creation of technical drawings. The approach is validated through an industrial case study involving configurable mechanical assemblies, where performance is evaluated using a quantitative time-based framework. The results show a reduction in total configuration time from 8 to 16 h to approximately 2&amp;amp;ndash;3 h, corresponding to an average time saving of 79% and a significant decrease in effective engineering workload. Assessment of scalability indicates that engineering effort becomes largely independent of the number of configurations, while economic evaluation indicates strong return-on-investment potential under typical industrial conditions. The findings confirm that systematic CAD automation can substantially improve productivity and consistency in configuration-driven design processes, provided that parametric models and data structures are properly standardised.</p>
	]]></content:encoded>

	<dc:title>A Generalized Framework for CAD Design Automation of Configurable Products Through CAD&amp;amp;ndash;PLM Integration with Quantitative Industrial Validation</dc:title>
			<dc:creator>Jakub Radkovsky</dc:creator>
			<dc:creator>Adam Becvar</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050146</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-17</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-17</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>146</prism:startingPage>
		<prism:doi>10.3390/automation7050146</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/146</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/145">

	<title>Automation, Vol. 7, Pages 145: Improved Siamese Network for High-Resolution Printed Surface Defect Detection Under Few-Shot Conditions</title>
	<link>https://www.mdpi.com/2673-4052/7/5/145</link>
	<description>Printed product surface defect inspection faces critical challenges in real industrial production, including insufficient detection accuracy, poor robustness, and scarcity of labeled defective samples. This paper proposes an improved Siamese network for high-resolution printed image defect detection to address the above limitations. Original 1300 &amp;amp;times; 460-pixel printed images are segmented into 64 &amp;amp;times; 64 patches to construct a training and testing dataset. We optimize the feature measurement strategy of contrastive loss and integrate a position matching module and multi-threshold evaluation strategy to balance detection precision and stability. Quantitative experiments demonstrate that the proposed method achieves an overall accuracy of 94.35%, a precision of 91.99%, a recall of 95.17%, and an F1-score of 93.55% at the fixed threshold of 0.45. Compared with mainstream lightweight detection models including YOLOv8Lite (You Only Look Once version 8 Lite) and MobileViTv2 (Mobile Vision Transformer version 2), our method maintains a stable F1-score advantage of 10.84&amp;amp;ndash;24.36% under optimal threshold settings. The proposed approach achieves outstanding performance and robustness in few-shot scenarios and can be deployed for automatic quality inspection of printed parts used in power communication equipment.</description>
	<pubDate>2026-09-17</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 145: Improved Siamese Network for High-Resolution Printed Surface Defect Detection Under Few-Shot Conditions</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/145">doi: 10.3390/automation7050145</a></p>
	<p>Authors:
		Renhao Zhu
		Chen Zhang
		</p>
	<p>Printed product surface defect inspection faces critical challenges in real industrial production, including insufficient detection accuracy, poor robustness, and scarcity of labeled defective samples. This paper proposes an improved Siamese network for high-resolution printed image defect detection to address the above limitations. Original 1300 &amp;amp;times; 460-pixel printed images are segmented into 64 &amp;amp;times; 64 patches to construct a training and testing dataset. We optimize the feature measurement strategy of contrastive loss and integrate a position matching module and multi-threshold evaluation strategy to balance detection precision and stability. Quantitative experiments demonstrate that the proposed method achieves an overall accuracy of 94.35%, a precision of 91.99%, a recall of 95.17%, and an F1-score of 93.55% at the fixed threshold of 0.45. Compared with mainstream lightweight detection models including YOLOv8Lite (You Only Look Once version 8 Lite) and MobileViTv2 (Mobile Vision Transformer version 2), our method maintains a stable F1-score advantage of 10.84&amp;amp;ndash;24.36% under optimal threshold settings. The proposed approach achieves outstanding performance and robustness in few-shot scenarios and can be deployed for automatic quality inspection of printed parts used in power communication equipment.</p>
	]]></content:encoded>

	<dc:title>Improved Siamese Network for High-Resolution Printed Surface Defect Detection Under Few-Shot Conditions</dc:title>
			<dc:creator>Renhao Zhu</dc:creator>
			<dc:creator>Chen Zhang</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050145</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-17</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-17</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>145</prism:startingPage>
		<prism:doi>10.3390/automation7050145</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/145</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/144">

	<title>Automation, Vol. 7, Pages 144: Multi-Station IoT Architecture for Distributed Thermal Monitoring and Fuzzy Heat-Risk Screening</title>
	<link>https://www.mdpi.com/2673-4052/7/5/144</link>
	<description>This paper presents a three-station Internet of Things (IoT) architecture for distributed thermal monitoring and local heater automation. The scope is deliberately limited to environmental monitoring, hysteresis-control characterization, and heat-risk screening; it is not presented as a complete thermal-comfort assessment or an occupational heat-stress instrument. Each workstation integrates an M5Stack AtomS3 Lite, a DHT22 temperature&amp;amp;ndash;humidity sensor installed at approximately 0.8 m above the floor, a relay, and a 1500 W, 110 V heater. Local automatic control uses a 20&amp;amp;ndash;25 &amp;amp;deg;C hysteresis band. Embedded acquisition/transmission was configured at 15 s, while accepted cloud records in the analyzed tests had median intervals of 30&amp;amp;ndash;31 s. Repeated automatic-control tests were identified for all three stations, yielding seven, five, and nine ON/OFF transition pairs for Stations 1, 2, and 3, respectively. The mean reported ON-transition temperatures were 19.94, 19.96, and 19.72 &amp;amp;deg;C, respectively, whereas mean reported OFF-transition temperatures were 25.20, 25.58, and 25.32 &amp;amp;deg;C. These results characterize the synchronized cloud-reported heater state, not an independently instrumented relay contact. Packet-level ESP-NOW logs were not available; consequently, packet loss and end-to-end latency are not inferred. In response to the thermal-assessment limitations, the fuzzy layer was simplified to Heat Index, a 5 min Heat-Index trend, and a 15 min hot-sample fraction. Estimated WBGT, Discomfort Index, PMV, and PPD were removed from the decision layer. The HeatRiskScore is therefore an operational heat-screening indicator rather than a comfort or occupational-risk metric. Against a simple instantaneous Heat-Index baseline, the fuzzy classifier reproduced all 18 stable/low-exposure synthetic scenarios and escalated 7/18 scenarios under high recent heat exposure and 9/18 under combined high exposure and rising trend, demonstrating the intended temporal added value without claiming external validation. A datasheet-bound uncertainty sensitivity and a 108-case PMV/PPD factorial analysis further quantify the limitations of low-cost sensing and assumed comfort inputs.</description>
	<pubDate>2026-09-16</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 144: Multi-Station IoT Architecture for Distributed Thermal Monitoring and Fuzzy Heat-Risk Screening</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/144">doi: 10.3390/automation7050144</a></p>
	<p>Authors:
		José Varela-Aldás
		Silvia Ayala-Trujillo
		Carolina Del-Valle-Soto
		</p>
	<p>This paper presents a three-station Internet of Things (IoT) architecture for distributed thermal monitoring and local heater automation. The scope is deliberately limited to environmental monitoring, hysteresis-control characterization, and heat-risk screening; it is not presented as a complete thermal-comfort assessment or an occupational heat-stress instrument. Each workstation integrates an M5Stack AtomS3 Lite, a DHT22 temperature&amp;amp;ndash;humidity sensor installed at approximately 0.8 m above the floor, a relay, and a 1500 W, 110 V heater. Local automatic control uses a 20&amp;amp;ndash;25 &amp;amp;deg;C hysteresis band. Embedded acquisition/transmission was configured at 15 s, while accepted cloud records in the analyzed tests had median intervals of 30&amp;amp;ndash;31 s. Repeated automatic-control tests were identified for all three stations, yielding seven, five, and nine ON/OFF transition pairs for Stations 1, 2, and 3, respectively. The mean reported ON-transition temperatures were 19.94, 19.96, and 19.72 &amp;amp;deg;C, respectively, whereas mean reported OFF-transition temperatures were 25.20, 25.58, and 25.32 &amp;amp;deg;C. These results characterize the synchronized cloud-reported heater state, not an independently instrumented relay contact. Packet-level ESP-NOW logs were not available; consequently, packet loss and end-to-end latency are not inferred. In response to the thermal-assessment limitations, the fuzzy layer was simplified to Heat Index, a 5 min Heat-Index trend, and a 15 min hot-sample fraction. Estimated WBGT, Discomfort Index, PMV, and PPD were removed from the decision layer. The HeatRiskScore is therefore an operational heat-screening indicator rather than a comfort or occupational-risk metric. Against a simple instantaneous Heat-Index baseline, the fuzzy classifier reproduced all 18 stable/low-exposure synthetic scenarios and escalated 7/18 scenarios under high recent heat exposure and 9/18 under combined high exposure and rising trend, demonstrating the intended temporal added value without claiming external validation. A datasheet-bound uncertainty sensitivity and a 108-case PMV/PPD factorial analysis further quantify the limitations of low-cost sensing and assumed comfort inputs.</p>
	]]></content:encoded>

	<dc:title>Multi-Station IoT Architecture for Distributed Thermal Monitoring and Fuzzy Heat-Risk Screening</dc:title>
			<dc:creator>José Varela-Aldás</dc:creator>
			<dc:creator>Silvia Ayala-Trujillo</dc:creator>
			<dc:creator>Carolina Del-Valle-Soto</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050144</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-16</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-16</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>144</prism:startingPage>
		<prism:doi>10.3390/automation7050144</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/144</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/143">

	<title>Automation, Vol. 7, Pages 143: Measurement and Simulation of Common-Mode Current for an Electric Vehicle Fast Charger</title>
	<link>https://www.mdpi.com/2673-4052/7/5/143</link>
	<description>Measurements and simulations of common-mode (CM) current in a commercial ABB TERRA 54CJ 50 kW electric vehicle fast charger are presented. Three-phase and neutral currents were measured using a Chauvin-Arnoux C.A 8336 power analyzer equipped with four MN93A current probes while charging a Hyundai Kona electric vehicle at approximately 50.5 kW. A reconstructed CM current was derived from the measured phase and neutral currents and analyzed in both the time and frequency domains. Fast Fourier Transform analysis showed that the dominant spectral component occurs at 50 Hz, with significant components below 2.5 kHz. An improved frequency-domain behavioral model was implemented in ADS using all measured harmonics. Comparison with measurements showed improved agreement relative to the previously published model, reducing average waveform errors for all phase currents and the neutral current. The results demonstrate the applicability of the proposed model for studying low-frequency common-mode current behavior in EV fast chargers and for supporting future EMC-related investigations.</description>
	<pubDate>2026-09-15</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 143: Measurement and Simulation of Common-Mode Current for an Electric Vehicle Fast Charger</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/143">doi: 10.3390/automation7050143</a></p>
	<p>Authors:
		Cosmin Gabriel Barancea
		Alexandru Gabriel Gheorghe
		Mihai Maricaru
		Mihăiță Gabriel Neacșu
		</p>
	<p>Measurements and simulations of common-mode (CM) current in a commercial ABB TERRA 54CJ 50 kW electric vehicle fast charger are presented. Three-phase and neutral currents were measured using a Chauvin-Arnoux C.A 8336 power analyzer equipped with four MN93A current probes while charging a Hyundai Kona electric vehicle at approximately 50.5 kW. A reconstructed CM current was derived from the measured phase and neutral currents and analyzed in both the time and frequency domains. Fast Fourier Transform analysis showed that the dominant spectral component occurs at 50 Hz, with significant components below 2.5 kHz. An improved frequency-domain behavioral model was implemented in ADS using all measured harmonics. Comparison with measurements showed improved agreement relative to the previously published model, reducing average waveform errors for all phase currents and the neutral current. The results demonstrate the applicability of the proposed model for studying low-frequency common-mode current behavior in EV fast chargers and for supporting future EMC-related investigations.</p>
	]]></content:encoded>

	<dc:title>Measurement and Simulation of Common-Mode Current for an Electric Vehicle Fast Charger</dc:title>
			<dc:creator>Cosmin Gabriel Barancea</dc:creator>
			<dc:creator>Alexandru Gabriel Gheorghe</dc:creator>
			<dc:creator>Mihai Maricaru</dc:creator>
			<dc:creator>Mihăiță Gabriel Neacșu</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050143</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-15</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-15</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Communication</prism:section>
	<prism:startingPage>143</prism:startingPage>
		<prism:doi>10.3390/automation7050143</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/143</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/142">

	<title>Automation, Vol. 7, Pages 142: GL-TGS: Guarded Learning-Based Time-Gap Supervision for Adaptive MPC in Perception-in-the-Loop Autonomous Highway Driving</title>
	<link>https://www.mdpi.com/2673-4052/7/5/142</link>
	<description>Adaptive cruise control (ACC) tracks a time-gap reference that trades efficiency against spacing safety; in practice this gap is fixed offline, although the safest choice depends on the closing speed and on the reliability of the perception that supplies the lead-vehicle state. We present GL-TGS-v2, a guarded, interpretable supervisor that selects the time-gap reference of an adaptive model predictive controller (MPC) online from tracked lead-vehicle kinematics. The supervisor is a decision-tree distillation of a calibrated adaptive expert, wrapped by a perception-aware safety shield; its only authority is the time-gap reference, so it retrofits onto an existing controller without re-opening the inner loop. We evaluate it in closed loop on a frozen stack that couples a YOLO11 detector, a joint probabilistic data association (JPDA) tracker, and the adaptive MPC, with documented runtime and per-run provenance gates. In a hard lead-braking scenario (ten runs per method), GL-TGS-v2 matches the expert at a matched mean time gap and improves minimum relative distance over fuzzy and fixed-gap baselines by 3.3&amp;amp;ndash;5.2 m, with 95% confidence intervals excluding zero, lower target loss, and no collisions. Benign testing shows no regression, while shield ablation characterizes the guard as an auditable protective override.</description>
	<pubDate>2026-09-14</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 142: GL-TGS: Guarded Learning-Based Time-Gap Supervision for Adaptive MPC in Perception-in-the-Loop Autonomous Highway Driving</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/142">doi: 10.3390/automation7050142</a></p>
	<p>Authors:
		Khaled Abd El Salam
		Hossam ElSayed
		Rehab F. Abdel-Kader
		Marwa Gamal
		</p>
	<p>Adaptive cruise control (ACC) tracks a time-gap reference that trades efficiency against spacing safety; in practice this gap is fixed offline, although the safest choice depends on the closing speed and on the reliability of the perception that supplies the lead-vehicle state. We present GL-TGS-v2, a guarded, interpretable supervisor that selects the time-gap reference of an adaptive model predictive controller (MPC) online from tracked lead-vehicle kinematics. The supervisor is a decision-tree distillation of a calibrated adaptive expert, wrapped by a perception-aware safety shield; its only authority is the time-gap reference, so it retrofits onto an existing controller without re-opening the inner loop. We evaluate it in closed loop on a frozen stack that couples a YOLO11 detector, a joint probabilistic data association (JPDA) tracker, and the adaptive MPC, with documented runtime and per-run provenance gates. In a hard lead-braking scenario (ten runs per method), GL-TGS-v2 matches the expert at a matched mean time gap and improves minimum relative distance over fuzzy and fixed-gap baselines by 3.3&amp;amp;ndash;5.2 m, with 95% confidence intervals excluding zero, lower target loss, and no collisions. Benign testing shows no regression, while shield ablation characterizes the guard as an auditable protective override.</p>
	]]></content:encoded>

	<dc:title>GL-TGS: Guarded Learning-Based Time-Gap Supervision for Adaptive MPC in Perception-in-the-Loop Autonomous Highway Driving</dc:title>
			<dc:creator>Khaled Abd El Salam</dc:creator>
			<dc:creator>Hossam ElSayed</dc:creator>
			<dc:creator>Rehab F. Abdel-Kader</dc:creator>
			<dc:creator>Marwa Gamal</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050142</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-14</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-14</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>142</prism:startingPage>
		<prism:doi>10.3390/automation7050142</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/142</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/141">

	<title>Automation, Vol. 7, Pages 141: Adaptive Fuzzy Logic-Based LED Power Control for Phaseolus vulgaris Cultivation Using Integrated Image Analysis and Illuminance Sensing</title>
	<link>https://www.mdpi.com/2673-4052/7/5/141</link>
	<description>This paper presents the design, implementation, and evaluation of a fuzzy logic-based adaptive lighting strategy for Phaseolus vulgaris cultivation. The objective of this study was to integrate an image-derived plant growth indicator obtained through projected leaf-area estimation into an adaptive pulse-width modulation framework. The proposed system used two input variables: the error between the desired and measured Photosynthetic Photon Flux Density (PPFD) and the estimated projected leaf area. PPFD was indirectly estimated from illuminance measurements acquired using a BH1750 sensor (ROHM Semiconductor, Kyoto, Japan) Mostrar m&amp;amp;aacute;s l&amp;amp;iacute;neas connected to an Arduino Uno microcontroller, while projected leaf area was obtained from RGB images captured with an Intel RealSense D435 camera and processed in MATLAB R2021a. The fuzzy inference system was implemented using the Mamdani approach and the Fuzzy Logic Designer application. The experimental results showed that the system regulated the PWM signal within a range of 17.00% to 87.85%, corresponding to a power range of 13.65 W to 34.58 W. The projected leaf-area indicator increased from 284.94 cm2 to 392.57 cm2 during the 21-day evaluation period, representing a relative increase of 37.77%. These findings provide preliminary evidence supporting the integration of image-derived projected leaf-area information, illuminance sensing, and fuzzy logic for adaptive lighting regulation in controlled-environment cultivation systems.</description>
	<pubDate>2026-09-11</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 141: Adaptive Fuzzy Logic-Based LED Power Control for Phaseolus vulgaris Cultivation Using Integrated Image Analysis and Illuminance Sensing</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/141">doi: 10.3390/automation7050141</a></p>
	<p>Authors:
		María J. Herrera-Hernández
		Marcos J. Villaseñor-Aguilar
		Carlos G. Manriquez-Padilla
		Adolfo R. Lopez
		Alejandro Barrientos-García
		Jose-Eleazar Peralta-Lopez
		</p>
	<p>This paper presents the design, implementation, and evaluation of a fuzzy logic-based adaptive lighting strategy for Phaseolus vulgaris cultivation. The objective of this study was to integrate an image-derived plant growth indicator obtained through projected leaf-area estimation into an adaptive pulse-width modulation framework. The proposed system used two input variables: the error between the desired and measured Photosynthetic Photon Flux Density (PPFD) and the estimated projected leaf area. PPFD was indirectly estimated from illuminance measurements acquired using a BH1750 sensor (ROHM Semiconductor, Kyoto, Japan) Mostrar m&amp;amp;aacute;s l&amp;amp;iacute;neas connected to an Arduino Uno microcontroller, while projected leaf area was obtained from RGB images captured with an Intel RealSense D435 camera and processed in MATLAB R2021a. The fuzzy inference system was implemented using the Mamdani approach and the Fuzzy Logic Designer application. The experimental results showed that the system regulated the PWM signal within a range of 17.00% to 87.85%, corresponding to a power range of 13.65 W to 34.58 W. The projected leaf-area indicator increased from 284.94 cm2 to 392.57 cm2 during the 21-day evaluation period, representing a relative increase of 37.77%. These findings provide preliminary evidence supporting the integration of image-derived projected leaf-area information, illuminance sensing, and fuzzy logic for adaptive lighting regulation in controlled-environment cultivation systems.</p>
	]]></content:encoded>

	<dc:title>Adaptive Fuzzy Logic-Based LED Power Control for Phaseolus vulgaris Cultivation Using Integrated Image Analysis and Illuminance Sensing</dc:title>
			<dc:creator>María J. Herrera-Hernández</dc:creator>
			<dc:creator>Marcos J. Villaseñor-Aguilar</dc:creator>
			<dc:creator>Carlos G. Manriquez-Padilla</dc:creator>
			<dc:creator>Adolfo R. Lopez</dc:creator>
			<dc:creator>Alejandro Barrientos-García</dc:creator>
			<dc:creator>Jose-Eleazar Peralta-Lopez</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050141</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-11</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-11</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>141</prism:startingPage>
		<prism:doi>10.3390/automation7050141</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/141</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/140">

	<title>Automation, Vol. 7, Pages 140: Intelligent Control Methods for Wheel-Slip Regulation in Automotive Braking Systems: A Systematic Review and Real-Time Embedded Feasibility Analysis</title>
	<link>https://www.mdpi.com/2673-4052/7/5/140</link>
	<description>This paper presents a systematic literature review of intelligent and advanced control methods for automotive wheel-slip regulation in anti-lock braking systems (ABS). Following a PRISMA-based workflow, records from IEEE Xplore, Scopus, and Engineering Village were searched for the 2014&amp;amp;ndash;2025 period. The search returned 5034 records; after removal of duplicates, patents, and out-of-range items, 1803 records were screened, 406 reports were assessed at full-text level, and 60 studies met the final eligibility criteria. The included studies were classified using a six-class taxonomy: fuzzy and neuro-fuzzy control, adaptive and self-tuning control, robust nonlinear control, predictive and optimization-based control, learning-assisted control, and hybrid or integrated control. The review shows that robust nonlinear and hybrid/integrated methods dominate the evidence base, while fuzzy and predictive methods remain important recurring families. Validation is still strongly simulation-weighted: many studies report braking-performance gains in slip tracking, stopping distance, chattering reduction, or road-friction robustness, but comparatively few provide hardware-in-the-loop, laboratory, or processor-level evidence. Only a very limited subset reports concrete embedded metrics such as execution time, sampling-period compliance, memory usage, or deadline margin. By combining systematic study selection, braking-specific control equations, primary-method classification, validation coding, embedded-implementation assessment, and comparative-scope analysis, this review identifies a persistent gap between algorithmic ABS performance and deployable real-time embedded feasibility. The findings motivate standardized benchmarking of intelligent braking controllers under common wheel-slip scenarios, common plant models, and common embedded timing metrics.</description>
	<pubDate>2026-09-10</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 140: Intelligent Control Methods for Wheel-Slip Regulation in Automotive Braking Systems: A Systematic Review and Real-Time Embedded Feasibility Analysis</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/140">doi: 10.3390/automation7050140</a></p>
	<p>Authors:
		Adnan Shaout
		Luis E. Castaneda-Trejo
		</p>
	<p>This paper presents a systematic literature review of intelligent and advanced control methods for automotive wheel-slip regulation in anti-lock braking systems (ABS). Following a PRISMA-based workflow, records from IEEE Xplore, Scopus, and Engineering Village were searched for the 2014&amp;amp;ndash;2025 period. The search returned 5034 records; after removal of duplicates, patents, and out-of-range items, 1803 records were screened, 406 reports were assessed at full-text level, and 60 studies met the final eligibility criteria. The included studies were classified using a six-class taxonomy: fuzzy and neuro-fuzzy control, adaptive and self-tuning control, robust nonlinear control, predictive and optimization-based control, learning-assisted control, and hybrid or integrated control. The review shows that robust nonlinear and hybrid/integrated methods dominate the evidence base, while fuzzy and predictive methods remain important recurring families. Validation is still strongly simulation-weighted: many studies report braking-performance gains in slip tracking, stopping distance, chattering reduction, or road-friction robustness, but comparatively few provide hardware-in-the-loop, laboratory, or processor-level evidence. Only a very limited subset reports concrete embedded metrics such as execution time, sampling-period compliance, memory usage, or deadline margin. By combining systematic study selection, braking-specific control equations, primary-method classification, validation coding, embedded-implementation assessment, and comparative-scope analysis, this review identifies a persistent gap between algorithmic ABS performance and deployable real-time embedded feasibility. The findings motivate standardized benchmarking of intelligent braking controllers under common wheel-slip scenarios, common plant models, and common embedded timing metrics.</p>
	]]></content:encoded>

	<dc:title>Intelligent Control Methods for Wheel-Slip Regulation in Automotive Braking Systems: A Systematic Review and Real-Time Embedded Feasibility Analysis</dc:title>
			<dc:creator>Adnan Shaout</dc:creator>
			<dc:creator>Luis E. Castaneda-Trejo</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050140</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-10</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-10</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Systematic Review</prism:section>
	<prism:startingPage>140</prism:startingPage>
		<prism:doi>10.3390/automation7050140</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/140</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/139">

	<title>Automation, Vol. 7, Pages 139: Adaptive Observer Design for Partially Measured Nonlinear Cascade Systems with Unknown Parameters</title>
	<link>https://www.mdpi.com/2673-4052/7/5/139</link>
	<description>This paper addresses the design of an adaptive observer for a class of nonlinear cascade systems with partially measured states and unknown constant parameters. The considered systems have a cascade structure involving an unmeasured-state subsystem and a measured-state subsystem. The unknown parameter vector enters the measured state dynamics through a known distribution matrix, while the unmeasured state dynamics are described by a lower triangular subsystem satisfying suitable conditions. An adaptive observer is proposed to reconstruct the unmeasured state variables and to compensate for the effect of the unknown parameters using only the measured output and the known input. Sufficient gain conditions are derived through a Lyapunov-based analysis to guarantee asymptotic convergence of the state estimation errors in the absence of uncertainties, while ensuring boundedness of the parameter estimation error. The effect of bounded model uncertainties is then analyzed, and a &amp;amp;sigma;-modified adaptive law is introduced to prevent parameter drift and guarantee uniform ultimate boundedness of the estimation errors. The proposed framework is illustrated through a reduced-order marine vessel model involving slowly varying environmental bias states and an unknown constant disturbance term. Numerical simulations illustrate the convergence properties and the robustness of the observer under bounded model uncertainties. Additional numerical tests with noisy output measurements are also reported to assess sensitivity to measurement noise.</description>
	<pubDate>2026-09-03</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 139: Adaptive Observer Design for Partially Measured Nonlinear Cascade Systems with Unknown Parameters</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/139">doi: 10.3390/automation7050139</a></p>
	<p>Authors:
		Francesco Pierri
		Graziano Carriero
		Monica Sileo
		Fabrizio Caccavale
		</p>
	<p>This paper addresses the design of an adaptive observer for a class of nonlinear cascade systems with partially measured states and unknown constant parameters. The considered systems have a cascade structure involving an unmeasured-state subsystem and a measured-state subsystem. The unknown parameter vector enters the measured state dynamics through a known distribution matrix, while the unmeasured state dynamics are described by a lower triangular subsystem satisfying suitable conditions. An adaptive observer is proposed to reconstruct the unmeasured state variables and to compensate for the effect of the unknown parameters using only the measured output and the known input. Sufficient gain conditions are derived through a Lyapunov-based analysis to guarantee asymptotic convergence of the state estimation errors in the absence of uncertainties, while ensuring boundedness of the parameter estimation error. The effect of bounded model uncertainties is then analyzed, and a &amp;amp;sigma;-modified adaptive law is introduced to prevent parameter drift and guarantee uniform ultimate boundedness of the estimation errors. The proposed framework is illustrated through a reduced-order marine vessel model involving slowly varying environmental bias states and an unknown constant disturbance term. Numerical simulations illustrate the convergence properties and the robustness of the observer under bounded model uncertainties. Additional numerical tests with noisy output measurements are also reported to assess sensitivity to measurement noise.</p>
	]]></content:encoded>

	<dc:title>Adaptive Observer Design for Partially Measured Nonlinear Cascade Systems with Unknown Parameters</dc:title>
			<dc:creator>Francesco Pierri</dc:creator>
			<dc:creator>Graziano Carriero</dc:creator>
			<dc:creator>Monica Sileo</dc:creator>
			<dc:creator>Fabrizio Caccavale</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050139</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-03</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-03</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>139</prism:startingPage>
		<prism:doi>10.3390/automation7050139</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/139</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/138">

	<title>Automation, Vol. 7, Pages 138: Towards Agentic Virtual Power Plants for Grid-Interactive Energy Communities: A Review-Informed Reference Architecture for Operational Flexibility Intelligence</title>
	<link>https://www.mdpi.com/2673-4052/7/5/138</link>
	<description>The growing deployment of distributed renewable generation, storage, electric vehicles, heat pumps, smart buildings, and controllable demand is expanding the flexibility available to local energy systems. Energy communities provide the governance context for collective participation and value creation, while community virtual power plants provide the operational mechanism for aggregating distributed resources and connecting them to local optimisation, flexibility markets, grid services, and resilience functions. This PRISMA-ScR-informed framework development study synthesises the literature from Scopus, Web of Science, and IEEE Xplore to examine how artificial intelligence supports community VPP operation, where agentic AI adds capabilities beyond established optimisation, reinforcement learning, and multi-agent systems, and which design requirements follow governed orchestration. The synthesis shows that current evidence is strongest for component-level forecasting, scheduling, bidding, adaptive control, distributed coordination, and digital-twin validation, whereas integrated agentic orchestration remains an emerging direction. Classical multi-agent systems already provide decentralised representation, communication, negotiation, and coordinated control; the additional role proposed for agentic AI is therefore narrower and concerns context-aware multi-step workflow orchestration, governed tool use, exception handling, grounded explanation, and bounded delegation across existing analytical and control services. The study introduces operational flexibility intelligence as the capability to transform potential distributed flexibility into deployable, authorised, market-, grid-, resilience-, and community-compatible action. It further develops a conceptual layered reference architecture in which agentic orchestration operates through governed tools and interfaces rather than bypassing validated resource controllers. Fairness, comfort, privacy, cybersecurity, resilience, auditability, and human-in-command authority are treated as cross-cutting operational constraints. The architecture defines a design and validation agenda rather than an empirically validated implementation.</description>
	<pubDate>2026-09-03</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 138: Towards Agentic Virtual Power Plants for Grid-Interactive Energy Communities: A Review-Informed Reference Architecture for Operational Flexibility Intelligence</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/138">doi: 10.3390/automation7050138</a></p>
	<p>Authors:
		Bo Nørregaard Jørgensen
		Zheng Grace Ma
		</p>
	<p>The growing deployment of distributed renewable generation, storage, electric vehicles, heat pumps, smart buildings, and controllable demand is expanding the flexibility available to local energy systems. Energy communities provide the governance context for collective participation and value creation, while community virtual power plants provide the operational mechanism for aggregating distributed resources and connecting them to local optimisation, flexibility markets, grid services, and resilience functions. This PRISMA-ScR-informed framework development study synthesises the literature from Scopus, Web of Science, and IEEE Xplore to examine how artificial intelligence supports community VPP operation, where agentic AI adds capabilities beyond established optimisation, reinforcement learning, and multi-agent systems, and which design requirements follow governed orchestration. The synthesis shows that current evidence is strongest for component-level forecasting, scheduling, bidding, adaptive control, distributed coordination, and digital-twin validation, whereas integrated agentic orchestration remains an emerging direction. Classical multi-agent systems already provide decentralised representation, communication, negotiation, and coordinated control; the additional role proposed for agentic AI is therefore narrower and concerns context-aware multi-step workflow orchestration, governed tool use, exception handling, grounded explanation, and bounded delegation across existing analytical and control services. The study introduces operational flexibility intelligence as the capability to transform potential distributed flexibility into deployable, authorised, market-, grid-, resilience-, and community-compatible action. It further develops a conceptual layered reference architecture in which agentic orchestration operates through governed tools and interfaces rather than bypassing validated resource controllers. Fairness, comfort, privacy, cybersecurity, resilience, auditability, and human-in-command authority are treated as cross-cutting operational constraints. The architecture defines a design and validation agenda rather than an empirically validated implementation.</p>
	]]></content:encoded>

	<dc:title>Towards Agentic Virtual Power Plants for Grid-Interactive Energy Communities: A Review-Informed Reference Architecture for Operational Flexibility Intelligence</dc:title>
			<dc:creator>Bo Nørregaard Jørgensen</dc:creator>
			<dc:creator>Zheng Grace Ma</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050138</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-03</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-03</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Review</prism:section>
	<prism:startingPage>138</prism:startingPage>
		<prism:doi>10.3390/automation7050138</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/138</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/137">

	<title>Automation, Vol. 7, Pages 137: Hybrid ISMC&amp;ndash;FTC Design with PSO Tuning for Finite-Time Stabilization of a 2-DOF Robotic Manipulator</title>
	<link>https://www.mdpi.com/2673-4052/7/5/137</link>
	<description>This paper proposes a hybrid integral sliding mode and finite-time control (ISMC&amp;amp;ndash;FTC) strategy for a two-degree-of-freedom (2DOF) robotic manipulator, targeting finite-time stability and high-precision elliptical trajectory tracking. To ensure practical deployment, the controller explicitly enforces actuator saturation limits, joint kinematic bounds, workspace constraints, and prescribed stabilization time requirements. Particle swarm optimization (PSO) is employed to tune the FTC parameters, minimizing convergence time while guaranteeing constraint satisfaction. By integrating ISMC&amp;amp;rsquo;s inherent robustness against matched disturbances with a PSO-optimized FTC, the scheme eliminates the reaching phase and ensures rapid, finite-time convergence. The simulation results demonstrate that the proposed approach significantly reduces stabilization time for both joints, maintains exceptionally low tracking errors, and enforces all physical constraints under bounded disturbances and model uncertainties through explicit saturation limits and sliding manifold invariance. These results validate the framework&amp;amp;rsquo;s effectiveness and highlight its potential for safety-critical, high-precision robotic applications requiring guaranteed finite-time performance. Robustness is further validated under simultaneous disturbances and uncertainties, where the controller maintains stable performance and consistently satisfies the required robust stability condition.</description>
	<pubDate>2026-09-01</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 137: Hybrid ISMC&amp;ndash;FTC Design with PSO Tuning for Finite-Time Stabilization of a 2-DOF Robotic Manipulator</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/137">doi: 10.3390/automation7050137</a></p>
	<p>Authors:
		Samara H. Al-dahlaki
		Safanah M. Raafat
		Shibly A. AI-Samarraie
		Amjad J. Humaidi
		</p>
	<p>This paper proposes a hybrid integral sliding mode and finite-time control (ISMC&amp;amp;ndash;FTC) strategy for a two-degree-of-freedom (2DOF) robotic manipulator, targeting finite-time stability and high-precision elliptical trajectory tracking. To ensure practical deployment, the controller explicitly enforces actuator saturation limits, joint kinematic bounds, workspace constraints, and prescribed stabilization time requirements. Particle swarm optimization (PSO) is employed to tune the FTC parameters, minimizing convergence time while guaranteeing constraint satisfaction. By integrating ISMC&amp;amp;rsquo;s inherent robustness against matched disturbances with a PSO-optimized FTC, the scheme eliminates the reaching phase and ensures rapid, finite-time convergence. The simulation results demonstrate that the proposed approach significantly reduces stabilization time for both joints, maintains exceptionally low tracking errors, and enforces all physical constraints under bounded disturbances and model uncertainties through explicit saturation limits and sliding manifold invariance. These results validate the framework&amp;amp;rsquo;s effectiveness and highlight its potential for safety-critical, high-precision robotic applications requiring guaranteed finite-time performance. Robustness is further validated under simultaneous disturbances and uncertainties, where the controller maintains stable performance and consistently satisfies the required robust stability condition.</p>
	]]></content:encoded>

	<dc:title>Hybrid ISMC&amp;amp;ndash;FTC Design with PSO Tuning for Finite-Time Stabilization of a 2-DOF Robotic Manipulator</dc:title>
			<dc:creator>Samara H. Al-dahlaki</dc:creator>
			<dc:creator>Safanah M. Raafat</dc:creator>
			<dc:creator>Shibly A. AI-Samarraie</dc:creator>
			<dc:creator>Amjad J. Humaidi</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050137</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-09-01</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-09-01</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>137</prism:startingPage>
		<prism:doi>10.3390/automation7050137</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/137</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/136">

	<title>Automation, Vol. 7, Pages 136: Substrate-Driven PLC Control (SD-PLC) Design for Fixed-Dome Biodigesters: A Rheology- and Energy-Constrained Methodology with a Guinea Pig Manure Co-Digestion Case Study</title>
	<link>https://www.mdpi.com/2673-4052/7/5/136</link>
	<description>The automation of biodigesters using programmable logic controllers (PLCs) is well established, yet mixing schedules are configured empirically, regardless of substrate characteristics. This study proposes a Substrate-Driven PLC Control (SD-PLC) methodology for fixed-dome biodigesters, in which every timing setpoint is derived from a measured substrate property: the homogenization pulse from the rheological mixing time (Metzner&amp;amp;ndash;Otto regime), the maintenance interval from a stratification and a substrate&amp;amp;ndash;biomass contact criterion, the duty-cycle ceiling from a net energy balance, and a safety gate from acid&amp;amp;ndash;base behavior (pH/EC). The methodology is instantiated on a 13.86 m3 fixed-dome biodigester in the Chill&amp;amp;oacute;n Valley (Lima, Peru), co-digesting organic waste and guinea pig manure (30:70; theoretical methane potential 371 mL CH4 g&amp;amp;minus;1 VS; field conversion &amp;amp;asymp; 21%). The impeller operates in the transitional regime (Re&amp;amp;asymp;0.9&amp;amp;ndash;2.6&amp;amp;times;103), the homogenization pulse is &amp;amp;asymp;22 min, and the reconciled duty cycle (&amp;amp;delta;&amp;amp;asymp;0.13) remains a factor of four below the energy ceiling (&amp;amp;delta;*&amp;amp;asymp;0.56). An influence &amp;amp;times; feasibility matrix identifies a low-cost sensor set (pH, electrical conductivity, temperature, level) that gates mixing. A reduced COD-based dynamic model shows that, under acid shock, continuous mixing suppresses methanogenesis through shear while the absence of mixing fails through contact deficit, so that only the stability-first strategy preserves the process. The principal contribution is the methodology itself&amp;amp;mdash;a reproducible mapping from substrate characterization to PLC timing design&amp;amp;mdash;rather than the hardware.</description>
	<pubDate>2026-08-31</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 136: Substrate-Driven PLC Control (SD-PLC) Design for Fixed-Dome Biodigesters: A Rheology- and Energy-Constrained Methodology with a Guinea Pig Manure Co-Digestion Case Study</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/136">doi: 10.3390/automation7050136</a></p>
	<p>Authors:
		Yoisdel Castillo Alvarez
		Anibal Salvador Wenceslao Ferro Diaz
		Deyvi Gair Baltazar Cari
		Reinier Jiménez Borges
		Carlos Diego Patiño Vidal
		Fanny Mabel Carhuancho León
		Romel Ángel Cárdenas Javier
		Roberto Pfuyo Muñoz
		</p>
	<p>The automation of biodigesters using programmable logic controllers (PLCs) is well established, yet mixing schedules are configured empirically, regardless of substrate characteristics. This study proposes a Substrate-Driven PLC Control (SD-PLC) methodology for fixed-dome biodigesters, in which every timing setpoint is derived from a measured substrate property: the homogenization pulse from the rheological mixing time (Metzner&amp;amp;ndash;Otto regime), the maintenance interval from a stratification and a substrate&amp;amp;ndash;biomass contact criterion, the duty-cycle ceiling from a net energy balance, and a safety gate from acid&amp;amp;ndash;base behavior (pH/EC). The methodology is instantiated on a 13.86 m3 fixed-dome biodigester in the Chill&amp;amp;oacute;n Valley (Lima, Peru), co-digesting organic waste and guinea pig manure (30:70; theoretical methane potential 371 mL CH4 g&amp;amp;minus;1 VS; field conversion &amp;amp;asymp; 21%). The impeller operates in the transitional regime (Re&amp;amp;asymp;0.9&amp;amp;ndash;2.6&amp;amp;times;103), the homogenization pulse is &amp;amp;asymp;22 min, and the reconciled duty cycle (&amp;amp;delta;&amp;amp;asymp;0.13) remains a factor of four below the energy ceiling (&amp;amp;delta;*&amp;amp;asymp;0.56). An influence &amp;amp;times; feasibility matrix identifies a low-cost sensor set (pH, electrical conductivity, temperature, level) that gates mixing. A reduced COD-based dynamic model shows that, under acid shock, continuous mixing suppresses methanogenesis through shear while the absence of mixing fails through contact deficit, so that only the stability-first strategy preserves the process. The principal contribution is the methodology itself&amp;amp;mdash;a reproducible mapping from substrate characterization to PLC timing design&amp;amp;mdash;rather than the hardware.</p>
	]]></content:encoded>

	<dc:title>Substrate-Driven PLC Control (SD-PLC) Design for Fixed-Dome Biodigesters: A Rheology- and Energy-Constrained Methodology with a Guinea Pig Manure Co-Digestion Case Study</dc:title>
			<dc:creator>Yoisdel Castillo Alvarez</dc:creator>
			<dc:creator>Anibal Salvador Wenceslao Ferro Diaz</dc:creator>
			<dc:creator>Deyvi Gair Baltazar Cari</dc:creator>
			<dc:creator>Reinier Jiménez Borges</dc:creator>
			<dc:creator>Carlos Diego Patiño Vidal</dc:creator>
			<dc:creator>Fanny Mabel Carhuancho León</dc:creator>
			<dc:creator>Romel Ángel Cárdenas Javier</dc:creator>
			<dc:creator>Roberto Pfuyo Muñoz</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050136</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-08-31</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-08-31</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>136</prism:startingPage>
		<prism:doi>10.3390/automation7050136</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/136</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/135">

	<title>Automation, Vol. 7, Pages 135: A Gated Recurrent Unit and Physics-Informed Neural Network-Based Method for Throughput Prediction of High-Pressure Grinding Rolls</title>
	<link>https://www.mdpi.com/2673-4052/7/5/135</link>
	<description>As an important part in the mining crushing and grinding circuit, the high-pressure grinding roll (HPGR) is essential for energy efficiency, cost reduction, quality improvement and efficiency gains; therefore, it plays a key role in sustaining stable production and intelligent control of mining operations. Its instantaneous throughput and processing capacity directly influence the efficiency of the comminution system, the compatibility of production scheduling, and overall energy consumption. Therefore, these metrics serve as important indicators for intelligent optimization and stable operation. However, it is a difficult task to predict the throughput of HPGRs by traditional purely data-driven models. The process is characterized by strong nonlinearity, significant time-lag effects, and limited physical consistency. To address these challenges, this work develops a tailored throughput prediction framework for HPGRs by combining Gated Recurrent Units with Physics-Informed Neural Networks (GRU-PINN), which integrates an HPGR-specific volumetric throughput mechanism as a dedicated physical constraint. This approach first exploits the GRU network&amp;amp;rsquo;s &amp;amp;ldquo;reset&amp;amp;rdquo; and &amp;amp;ldquo;update&amp;amp;rdquo; gates to selectively filter historical operating information while adaptively updating the current process features. This allows the model to better capture complex long-term temporal dependencies in the production data. Additionally, based on volumetric analysis and the principles of bed comminution, the HPGR throughput formula is incorporated into the loss function as a physical constraint. Together, these components establish a joint optimization framework that combines data-driven learning with physical constraints to correct prediction biases generated by the neural network during abrupt changes in operating conditions. The proposed GRU-PINN model was validated using real operational data collected from an industrial mining site. The results show that the proposed framework significantly improves the physical consistency of HPGR throughput predictions and, at the same time, effectively corrects prediction biases, which are commonly observed in conventional purely data-driven models under complex operating conditions. It also exhibits improved robustness and lower inference latency. Compared with other benchmark models, the proposed method displays superior overall performance across key evaluation metrics, including root mean square error (RMSE), mean absolute error (MAE), prediction accuracy, and inference speed. These findings confirm that the proposed method greatly enhances the physical interpretability of the prediction model without compromising accuracy. Consequently, it lays a solid foundation for process parameter optimization and intelligent control of HPGR system.</description>
	<pubDate>2026-08-28</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 135: A Gated Recurrent Unit and Physics-Informed Neural Network-Based Method for Throughput Prediction of High-Pressure Grinding Rolls</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/135">doi: 10.3390/automation7050135</a></p>
	<p>Authors:
		Wenchao Yang
		Shihao Liu
		Junlin Zeng
		Rongchang Li
		Xiaoyu Wen
		Yuyan Zhang
		Yuanji Liang
		</p>
	<p>As an important part in the mining crushing and grinding circuit, the high-pressure grinding roll (HPGR) is essential for energy efficiency, cost reduction, quality improvement and efficiency gains; therefore, it plays a key role in sustaining stable production and intelligent control of mining operations. Its instantaneous throughput and processing capacity directly influence the efficiency of the comminution system, the compatibility of production scheduling, and overall energy consumption. Therefore, these metrics serve as important indicators for intelligent optimization and stable operation. However, it is a difficult task to predict the throughput of HPGRs by traditional purely data-driven models. The process is characterized by strong nonlinearity, significant time-lag effects, and limited physical consistency. To address these challenges, this work develops a tailored throughput prediction framework for HPGRs by combining Gated Recurrent Units with Physics-Informed Neural Networks (GRU-PINN), which integrates an HPGR-specific volumetric throughput mechanism as a dedicated physical constraint. This approach first exploits the GRU network&amp;amp;rsquo;s &amp;amp;ldquo;reset&amp;amp;rdquo; and &amp;amp;ldquo;update&amp;amp;rdquo; gates to selectively filter historical operating information while adaptively updating the current process features. This allows the model to better capture complex long-term temporal dependencies in the production data. Additionally, based on volumetric analysis and the principles of bed comminution, the HPGR throughput formula is incorporated into the loss function as a physical constraint. Together, these components establish a joint optimization framework that combines data-driven learning with physical constraints to correct prediction biases generated by the neural network during abrupt changes in operating conditions. The proposed GRU-PINN model was validated using real operational data collected from an industrial mining site. The results show that the proposed framework significantly improves the physical consistency of HPGR throughput predictions and, at the same time, effectively corrects prediction biases, which are commonly observed in conventional purely data-driven models under complex operating conditions. It also exhibits improved robustness and lower inference latency. Compared with other benchmark models, the proposed method displays superior overall performance across key evaluation metrics, including root mean square error (RMSE), mean absolute error (MAE), prediction accuracy, and inference speed. These findings confirm that the proposed method greatly enhances the physical interpretability of the prediction model without compromising accuracy. Consequently, it lays a solid foundation for process parameter optimization and intelligent control of HPGR system.</p>
	]]></content:encoded>

	<dc:title>A Gated Recurrent Unit and Physics-Informed Neural Network-Based Method for Throughput Prediction of High-Pressure Grinding Rolls</dc:title>
			<dc:creator>Wenchao Yang</dc:creator>
			<dc:creator>Shihao Liu</dc:creator>
			<dc:creator>Junlin Zeng</dc:creator>
			<dc:creator>Rongchang Li</dc:creator>
			<dc:creator>Xiaoyu Wen</dc:creator>
			<dc:creator>Yuyan Zhang</dc:creator>
			<dc:creator>Yuanji Liang</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050135</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-08-28</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-08-28</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>135</prism:startingPage>
		<prism:doi>10.3390/automation7050135</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/135</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/134">

	<title>Automation, Vol. 7, Pages 134: Robust Fractional-Order Control of Vehicle Platoon Under Actuator and Communication Delay Intervals</title>
	<link>https://www.mdpi.com/2673-4052/7/5/134</link>
	<description>Stabilisation of autonomous vehicle platoons becomes challenging in the presence of uncertain delays in actuator dynamics and vehicle-to-vehicle communication. This paper presents a distributed fractional-order proportional derivative (FOPD) control protocol for a platoon operating under various communication topologies, with bounded uncertainties in actuator lag and communication delay, respectively. It builds on an auxiliary function-based robust stability approach that provides a geometric interpretation of the delay-uncertain characteristic family and eliminates the need for exhaustive parameter gridding. This is followed by a numerical design procedure that evaluates the derived robust stability conditions on explicitly stated frequency grids. Numerical analysis of heterogeneous platoons indicates that the FOPD strategy produces bounded stable responses for the sampled uncertainty realisations and exhibits improved tracking behaviour compared with the integer-order PD baseline using the same numerical values of KP and KD. It also provides better damping of oscillations and reduced tracking errors under uncertain operating conditions.</description>
	<pubDate>2026-08-26</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 134: Robust Fractional-Order Control of Vehicle Platoon Under Actuator and Communication Delay Intervals</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/134">doi: 10.3390/automation7050134</a></p>
	<p>Authors:
		Majid Ghorbani
		Omar Hanif
		Patrick Gruber
		Aldo Sorniotti
		Komeil Nosrati
		Aleksei Tepljakov
		Eduard Petlenkov
		Umberto Montanaro
		</p>
	<p>Stabilisation of autonomous vehicle platoons becomes challenging in the presence of uncertain delays in actuator dynamics and vehicle-to-vehicle communication. This paper presents a distributed fractional-order proportional derivative (FOPD) control protocol for a platoon operating under various communication topologies, with bounded uncertainties in actuator lag and communication delay, respectively. It builds on an auxiliary function-based robust stability approach that provides a geometric interpretation of the delay-uncertain characteristic family and eliminates the need for exhaustive parameter gridding. This is followed by a numerical design procedure that evaluates the derived robust stability conditions on explicitly stated frequency grids. Numerical analysis of heterogeneous platoons indicates that the FOPD strategy produces bounded stable responses for the sampled uncertainty realisations and exhibits improved tracking behaviour compared with the integer-order PD baseline using the same numerical values of KP and KD. It also provides better damping of oscillations and reduced tracking errors under uncertain operating conditions.</p>
	]]></content:encoded>

	<dc:title>Robust Fractional-Order Control of Vehicle Platoon Under Actuator and Communication Delay Intervals</dc:title>
			<dc:creator>Majid Ghorbani</dc:creator>
			<dc:creator>Omar Hanif</dc:creator>
			<dc:creator>Patrick Gruber</dc:creator>
			<dc:creator>Aldo Sorniotti</dc:creator>
			<dc:creator>Komeil Nosrati</dc:creator>
			<dc:creator>Aleksei Tepljakov</dc:creator>
			<dc:creator>Eduard Petlenkov</dc:creator>
			<dc:creator>Umberto Montanaro</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050134</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-08-26</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-08-26</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>134</prism:startingPage>
		<prism:doi>10.3390/automation7050134</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/134</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/5/133">

	<title>Automation, Vol. 7, Pages 133: Modeling and RBFNN-AMSC Tracking Control of a Cable-Driven Underwater Vehicle with Unknown Disturbances</title>
	<link>https://www.mdpi.com/2673-4052/7/5/133</link>
	<description>To complete scientific experiments on an underwater tension leg platform, a new cable-driven underwater vehicle is proposed, which is subjected to not only unknown underwater disturbances but also time-varying nonlinear cable tractions. To achieve displacement tracking control despite the high-order nonlinearities and matched and mismatched uncertainties with unknown upper bounds, a radial basis function neural network-based adaptive multiple-surface sliding control strategy (RBFNN-AMSC) is proposed. Utilizing a backstepping design procedure and the Lyapunov approach, the system is decomposed into six subsystems, and the stability is ensured. By employing multiple-surface sliding mode control and exponential reaching law design, the robustness and convergence rate of each subsystem are improved. Moreover, with an adaptive radial basis function neural network, the influences of matched and mismatched uncertainties are compensated, which improves the system anti-jamming capability and avoids the &amp;amp;ldquo;differential explosion&amp;amp;rdquo; problem. To verify the effectiveness of the proposed approach, numerical simulations are carried out under different conditions, which show that the proposed control strategy can achieve accurate displacement tracking control regardless of complex nonlinear dynamics and various unknown external disturbances.</description>
	<pubDate>2026-08-25</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 133: Modeling and RBFNN-AMSC Tracking Control of a Cable-Driven Underwater Vehicle with Unknown Disturbances</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/5/133">doi: 10.3390/automation7050133</a></p>
	<p>Authors:
		Kan Xu
		Yingkai Xia
		</p>
	<p>To complete scientific experiments on an underwater tension leg platform, a new cable-driven underwater vehicle is proposed, which is subjected to not only unknown underwater disturbances but also time-varying nonlinear cable tractions. To achieve displacement tracking control despite the high-order nonlinearities and matched and mismatched uncertainties with unknown upper bounds, a radial basis function neural network-based adaptive multiple-surface sliding control strategy (RBFNN-AMSC) is proposed. Utilizing a backstepping design procedure and the Lyapunov approach, the system is decomposed into six subsystems, and the stability is ensured. By employing multiple-surface sliding mode control and exponential reaching law design, the robustness and convergence rate of each subsystem are improved. Moreover, with an adaptive radial basis function neural network, the influences of matched and mismatched uncertainties are compensated, which improves the system anti-jamming capability and avoids the &amp;amp;ldquo;differential explosion&amp;amp;rdquo; problem. To verify the effectiveness of the proposed approach, numerical simulations are carried out under different conditions, which show that the proposed control strategy can achieve accurate displacement tracking control regardless of complex nonlinear dynamics and various unknown external disturbances.</p>
	]]></content:encoded>

	<dc:title>Modeling and RBFNN-AMSC Tracking Control of a Cable-Driven Underwater Vehicle with Unknown Disturbances</dc:title>
			<dc:creator>Kan Xu</dc:creator>
			<dc:creator>Yingkai Xia</dc:creator>
		<dc:identifier>doi: 10.3390/automation7050133</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-08-25</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-08-25</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>5</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>133</prism:startingPage>
		<prism:doi>10.3390/automation7050133</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/5/133</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/132">

	<title>Automation, Vol. 7, Pages 132: Automated Risk Assessment and Control Framework for Feed Production in Digital Agroengineering Systems</title>
	<link>https://www.mdpi.com/2673-4052/7/4/132</link>
	<description>Digital transformation in agricultural production requires automated approaches for monitoring, risk assessment, and decision support under uncertainty. This study proposes an automated risk assessment and control framework for feed production in digital agroengineering systems. The framework is designed for the quantitative evaluation of producer and consumer risks arising during the control of key feed-quality parameters, using the case of feed production for cattle in the OHMK agricultural holding. The proposed approach integrates probabilistic modeling, simulation-based risk estimation, fuzzy logic, expert evaluation, and a multi-agent representation of agroengineering processes. A three-dimensional risk model is developed to represent producer risk, consumer risk, and actuarial risk as interconnected components of a digital control environment. In addition, a fuzzy model is introduced to assess the robustness and digital maturity of management functions, including organization, planning, motivation, and control. Computer experiments based on statistical data for crude protein content in silage demonstrate that control risks depend nonlinearly on measurement uncertainty, parameter variability, and normative thresholds. In the analyzed single-indicator case study, the arithmetic mean of crude protein content in silage was 7.5% of dry matter, the standard deviation was 0.5, and the Weibull approximation parameters were &amp;amp;alpha; = 1.0, &amp;amp;beta; = 2.5, and &amp;amp;gamma; = 6.0. Under the most sensitive normative threshold scenario, producer risk increased to approximately 25%, while consumer risk showed a lower but nonlinear increase with measurement uncertainty. The results show that producer risk may reach significant levels when measurement uncertainty becomes comparable with the variability of the controlled parameter. The proposed framework can serve as a computational basis for future automated monitoring, risk-aware control, and decision-support systems in Industry 4.0-oriented agricultural production.</description>
	<pubDate>2026-08-20</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 132: Automated Risk Assessment and Control Framework for Feed Production in Digital Agroengineering Systems</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/132">doi: 10.3390/automation7040132</a></p>
	<p>Authors:
		Farid Abitaev
		Bagdat Azamatov
		Suresh Alapati
		Vyacheslav Kornev
		Rustam Zhanbosinov
		Karlygash Alibekkyzy
		Madina Bazarova
		</p>
	<p>Digital transformation in agricultural production requires automated approaches for monitoring, risk assessment, and decision support under uncertainty. This study proposes an automated risk assessment and control framework for feed production in digital agroengineering systems. The framework is designed for the quantitative evaluation of producer and consumer risks arising during the control of key feed-quality parameters, using the case of feed production for cattle in the OHMK agricultural holding. The proposed approach integrates probabilistic modeling, simulation-based risk estimation, fuzzy logic, expert evaluation, and a multi-agent representation of agroengineering processes. A three-dimensional risk model is developed to represent producer risk, consumer risk, and actuarial risk as interconnected components of a digital control environment. In addition, a fuzzy model is introduced to assess the robustness and digital maturity of management functions, including organization, planning, motivation, and control. Computer experiments based on statistical data for crude protein content in silage demonstrate that control risks depend nonlinearly on measurement uncertainty, parameter variability, and normative thresholds. In the analyzed single-indicator case study, the arithmetic mean of crude protein content in silage was 7.5% of dry matter, the standard deviation was 0.5, and the Weibull approximation parameters were &amp;amp;alpha; = 1.0, &amp;amp;beta; = 2.5, and &amp;amp;gamma; = 6.0. Under the most sensitive normative threshold scenario, producer risk increased to approximately 25%, while consumer risk showed a lower but nonlinear increase with measurement uncertainty. The results show that producer risk may reach significant levels when measurement uncertainty becomes comparable with the variability of the controlled parameter. The proposed framework can serve as a computational basis for future automated monitoring, risk-aware control, and decision-support systems in Industry 4.0-oriented agricultural production.</p>
	]]></content:encoded>

	<dc:title>Automated Risk Assessment and Control Framework for Feed Production in Digital Agroengineering Systems</dc:title>
			<dc:creator>Farid Abitaev</dc:creator>
			<dc:creator>Bagdat Azamatov</dc:creator>
			<dc:creator>Suresh Alapati</dc:creator>
			<dc:creator>Vyacheslav Kornev</dc:creator>
			<dc:creator>Rustam Zhanbosinov</dc:creator>
			<dc:creator>Karlygash Alibekkyzy</dc:creator>
			<dc:creator>Madina Bazarova</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040132</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-08-20</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-08-20</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>132</prism:startingPage>
		<prism:doi>10.3390/automation7040132</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/132</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/131">

	<title>Automation, Vol. 7, Pages 131: YOLOv11&amp;ndash;BiFPN&amp;ndash;DAAF: An Object Detection Framework for Automated Surface Inspection of Balsa Wood Panels</title>
	<link>https://www.mdpi.com/2673-4052/7/4/131</link>
	<description>Automated surface inspection of balsa wood panels is challenging because defects may be small, elongated, weakly contrasted, or visually similar to natural grain patterns. This study proposes YOLOv11&amp;amp;ndash;BiFPN&amp;amp;ndash;DAAF, an enhanced object-detection architecture that combines bidirectional multi-scale feature fusion with adaptive dual-attention feature refinement. The task was formulated as single-class detection, with all anomalous surface regions labeled as Defect. The dataset comprised 508 manually annotated RGB images, including independent internal and external production test sets. A preliminary screening identified YOLOv11-m512 as the reference configuration, followed by a controlled 2&amp;amp;times;2 factorial ablation comprising the baseline, BiFPN, DAAF, and their combined integration. Each configuration was trained using five independent random seeds under identical experimental conditions. On the validation set, the combined architecture achieved a precision of 0.893&amp;amp;plusmn;0.004, recall of 0.848&amp;amp;plusmn;0.006, mAP@0.5 of 0.897&amp;amp;plusmn;0.004, and mAP@0.5:0.95 of 0.389&amp;amp;plusmn;0.004. Relative to the baseline, the largest improvement was obtained for mAP@0.5:0.95, with a relative gain of 9.89%, indicating improved localization under stricter IoU thresholds. The improvement was retained on the independent internal test set, where the proposed architecture reached mAP@0.5 and mAP@0.5:0.95 values of 0.892&amp;amp;plusmn;0.005 and 0.384&amp;amp;plusmn;0.006, respectively. On the external production test set, the corresponding values were 0.865&amp;amp;plusmn;0.007 and 0.358&amp;amp;plusmn;0.008, representing absolute improvements of 0.034 and 0.042 over the original YOLOv11 baseline. Under the matched experimental protocol, YOLOv11&amp;amp;ndash;BiFPN&amp;amp;ndash;DAAF also achieved the highest principal detection metrics among the evaluated representative detectors. Although BiFPN and DAAF introduced a moderate computational overhead, the architecture maintained an inference time of 9.6&amp;amp;plusmn;0.3 ms per image. These findings support the potential of the proposed architecture for automated balsa wood panel inspection, while broader multi-site and hardware-level validation remains necessary before large-scale industrial deployment.</description>
	<pubDate>2026-08-17</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 131: YOLOv11&amp;ndash;BiFPN&amp;ndash;DAAF: An Object Detection Framework for Automated Surface Inspection of Balsa Wood Panels</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/131">doi: 10.3390/automation7040131</a></p>
	<p>Authors:
		Cristian Zambrano-Vega
		Washington Chiriboga-Casanova
		Byron Oviedo
		Efraín Díaz-Macías
		Edgar Suárez Bardelline
		</p>
	<p>Automated surface inspection of balsa wood panels is challenging because defects may be small, elongated, weakly contrasted, or visually similar to natural grain patterns. This study proposes YOLOv11&amp;amp;ndash;BiFPN&amp;amp;ndash;DAAF, an enhanced object-detection architecture that combines bidirectional multi-scale feature fusion with adaptive dual-attention feature refinement. The task was formulated as single-class detection, with all anomalous surface regions labeled as Defect. The dataset comprised 508 manually annotated RGB images, including independent internal and external production test sets. A preliminary screening identified YOLOv11-m512 as the reference configuration, followed by a controlled 2&amp;amp;times;2 factorial ablation comprising the baseline, BiFPN, DAAF, and their combined integration. Each configuration was trained using five independent random seeds under identical experimental conditions. On the validation set, the combined architecture achieved a precision of 0.893&amp;amp;plusmn;0.004, recall of 0.848&amp;amp;plusmn;0.006, mAP@0.5 of 0.897&amp;amp;plusmn;0.004, and mAP@0.5:0.95 of 0.389&amp;amp;plusmn;0.004. Relative to the baseline, the largest improvement was obtained for mAP@0.5:0.95, with a relative gain of 9.89%, indicating improved localization under stricter IoU thresholds. The improvement was retained on the independent internal test set, where the proposed architecture reached mAP@0.5 and mAP@0.5:0.95 values of 0.892&amp;amp;plusmn;0.005 and 0.384&amp;amp;plusmn;0.006, respectively. On the external production test set, the corresponding values were 0.865&amp;amp;plusmn;0.007 and 0.358&amp;amp;plusmn;0.008, representing absolute improvements of 0.034 and 0.042 over the original YOLOv11 baseline. Under the matched experimental protocol, YOLOv11&amp;amp;ndash;BiFPN&amp;amp;ndash;DAAF also achieved the highest principal detection metrics among the evaluated representative detectors. Although BiFPN and DAAF introduced a moderate computational overhead, the architecture maintained an inference time of 9.6&amp;amp;plusmn;0.3 ms per image. These findings support the potential of the proposed architecture for automated balsa wood panel inspection, while broader multi-site and hardware-level validation remains necessary before large-scale industrial deployment.</p>
	]]></content:encoded>

	<dc:title>YOLOv11&amp;amp;ndash;BiFPN&amp;amp;ndash;DAAF: An Object Detection Framework for Automated Surface Inspection of Balsa Wood Panels</dc:title>
			<dc:creator>Cristian Zambrano-Vega</dc:creator>
			<dc:creator>Washington Chiriboga-Casanova</dc:creator>
			<dc:creator>Byron Oviedo</dc:creator>
			<dc:creator>Efraín Díaz-Macías</dc:creator>
			<dc:creator>Edgar Suárez Bardelline</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040131</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-08-17</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-08-17</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>131</prism:startingPage>
		<prism:doi>10.3390/automation7040131</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/131</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/130">

	<title>Automation, Vol. 7, Pages 130: Vision-Map Fusion Multi-Object Tracking at Complex Intersections Using HD Map Priors and Nonlinear Filtering</title>
	<link>https://www.mdpi.com/2673-4052/7/4/130</link>
	<description>Accurate multi-object tracking and metric localization support traffic monitoring and cooperative intelligent transportation at complex intersections. This study presents a fixed-camera vision-map fusion framework that addresses two practical difficulties: axis-aligned boxes poorly represent turning vehicles, and unconstrained image-plane tracking can produce physically implausible trajectories. A map-aided frontend first generates candidate detections using improved You Only Look Once version 8 nano (YOLOv8n) horizontal bounding box (HBB) branch and an improved YOLOv8 oriented bounding box (OBB) branch. A high-definition (HD) map selector then retains the candidate geometry consistent with the straight-driving or turning region and converts it into a unified detection record. The selected reference point is projected to the ground plane through an offline-estimated homography, whereas the appearance feature bypasses the homography and is passed directly to the association stage. The tracking backend uses a 12-dimensional joint image/metric state, symmetric central-difference evaluations of the process and measurement functions, appearance-motion association, and a feasible-road projection derived from HD-map lane polygons. On the evaluated public sequences, the complete configuration achieved a multiple object tracking accuracy (MOTA) of 74.5%, an identification F1 score (IDF1) of 82.6%, 614 identity switches, and a throughput of 26.8 frames per second (FPS) on an RTX 4090 workstation. In a descriptive Vehicle-in-the-Loop case study involving one instrumented vehicle at one intersection, the overall localization mean absolute error (MAE) was 0.180 m, compared with 0.208 m for the baseline end-to-end configuration. These results indicate the feasibility of combining branch-specific vehicle geometry with map-constrained tracking; controlled same-detector comparisons, repeated multi-vehicle trials, and embedded-device latency and power profiling remain necessary for broader claims.</description>
	<pubDate>2026-08-16</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 130: Vision-Map Fusion Multi-Object Tracking at Complex Intersections Using HD Map Priors and Nonlinear Filtering</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/130">doi: 10.3390/automation7040130</a></p>
	<p>Authors:
		Dezheng Ma
		Lan Tang
		</p>
	<p>Accurate multi-object tracking and metric localization support traffic monitoring and cooperative intelligent transportation at complex intersections. This study presents a fixed-camera vision-map fusion framework that addresses two practical difficulties: axis-aligned boxes poorly represent turning vehicles, and unconstrained image-plane tracking can produce physically implausible trajectories. A map-aided frontend first generates candidate detections using improved You Only Look Once version 8 nano (YOLOv8n) horizontal bounding box (HBB) branch and an improved YOLOv8 oriented bounding box (OBB) branch. A high-definition (HD) map selector then retains the candidate geometry consistent with the straight-driving or turning region and converts it into a unified detection record. The selected reference point is projected to the ground plane through an offline-estimated homography, whereas the appearance feature bypasses the homography and is passed directly to the association stage. The tracking backend uses a 12-dimensional joint image/metric state, symmetric central-difference evaluations of the process and measurement functions, appearance-motion association, and a feasible-road projection derived from HD-map lane polygons. On the evaluated public sequences, the complete configuration achieved a multiple object tracking accuracy (MOTA) of 74.5%, an identification F1 score (IDF1) of 82.6%, 614 identity switches, and a throughput of 26.8 frames per second (FPS) on an RTX 4090 workstation. In a descriptive Vehicle-in-the-Loop case study involving one instrumented vehicle at one intersection, the overall localization mean absolute error (MAE) was 0.180 m, compared with 0.208 m for the baseline end-to-end configuration. These results indicate the feasibility of combining branch-specific vehicle geometry with map-constrained tracking; controlled same-detector comparisons, repeated multi-vehicle trials, and embedded-device latency and power profiling remain necessary for broader claims.</p>
	]]></content:encoded>

	<dc:title>Vision-Map Fusion Multi-Object Tracking at Complex Intersections Using HD Map Priors and Nonlinear Filtering</dc:title>
			<dc:creator>Dezheng Ma</dc:creator>
			<dc:creator>Lan Tang</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040130</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-08-16</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-08-16</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>130</prism:startingPage>
		<prism:doi>10.3390/automation7040130</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/130</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/129">

	<title>Automation, Vol. 7, Pages 129: Run-Level Evaluation of a Confidence-Gated Kalman Lane-Keeping Architecture for a 1:10-Scale Autonomous Vehicle</title>
	<link>https://www.mdpi.com/2673-4052/7/4/129</link>
	<description>Lane keeping under degraded visual confidence remains challenging because most existing approaches focus either on improving lane-feature extraction or on estimating vehicle motion, while giving less attention to how unreliable visual measurements should modify the estimator&amp;amp;ndash;controller interaction in a physical closed-loop system. This paper presents a confidence-gated Kalman lane-keeping architecture for a 1:10-scale autonomous vehicle. The methodological contribution lies in the direct coupling of visual confidence, state estimation, and steering control: unreliable lane measurements are down-weighted through confidence-dependent measurement noise, while the propagated lane-relative state remains available to the controller. The primary experimental unit is the run, defined as one logged lap under one control configuration; frame-level summaries are retained only as historical reproducibility material. In the run-level comparison, the autonomous vision, inertial measurement unit (IMU)-feedforward, and Kalman filter (KF) group KF_G1&amp;amp;mdash;the first inferential KF generation&amp;amp;mdash;had lower mean absolute error than the human baseline, while vision and KF_G1 had overlapping run-level confidence intervals for absolute error. KF_G1 shifted the mean signed bias closer to the lane reference than the vision and IMU groups, but with higher run-level spread than vision. KF_G2, the second observational KF generation, is reported only as an observational generation comparison, so no causal claim is made for its estimator&amp;amp;ndash;controller correction weight Ks=0.15 setting. KF_G3, the single descriptive adverse-condition KF run, is reported without an estimable confidence interval or population-level adverse-illumination inference. A further limitation is that the inertial prediction pathway was inactive in the logged Kalman-filter runs because the inertial coupling coefficient &amp;amp;alpha;=0. The main contribution is a reproducible run-level evaluation of confidence-gated estimator&amp;amp;ndash;controller coupling that distinguishes supported evidence from observational and descriptive cases.</description>
	<pubDate>2026-08-13</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 129: Run-Level Evaluation of a Confidence-Gated Kalman Lane-Keeping Architecture for a 1:10-Scale Autonomous Vehicle</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/129">doi: 10.3390/automation7040129</a></p>
	<p>Authors:
		Rafael Reveles-Martínez
		Hamurabi Gamboa-Rosales
		Huizilopoztli Luna-García
		Erika Sánchez-Femat
		Javier Saldívar-Pérez
		Flabio D. Mirelez-Delgado
		Umanel A. Hernández-González
		Carlos E. Galván-Tejada
		Jorge I. Galván-Tejada
		José M. Celaya-Padilla
		</p>
	<p>Lane keeping under degraded visual confidence remains challenging because most existing approaches focus either on improving lane-feature extraction or on estimating vehicle motion, while giving less attention to how unreliable visual measurements should modify the estimator&amp;amp;ndash;controller interaction in a physical closed-loop system. This paper presents a confidence-gated Kalman lane-keeping architecture for a 1:10-scale autonomous vehicle. The methodological contribution lies in the direct coupling of visual confidence, state estimation, and steering control: unreliable lane measurements are down-weighted through confidence-dependent measurement noise, while the propagated lane-relative state remains available to the controller. The primary experimental unit is the run, defined as one logged lap under one control configuration; frame-level summaries are retained only as historical reproducibility material. In the run-level comparison, the autonomous vision, inertial measurement unit (IMU)-feedforward, and Kalman filter (KF) group KF_G1&amp;amp;mdash;the first inferential KF generation&amp;amp;mdash;had lower mean absolute error than the human baseline, while vision and KF_G1 had overlapping run-level confidence intervals for absolute error. KF_G1 shifted the mean signed bias closer to the lane reference than the vision and IMU groups, but with higher run-level spread than vision. KF_G2, the second observational KF generation, is reported only as an observational generation comparison, so no causal claim is made for its estimator&amp;amp;ndash;controller correction weight Ks=0.15 setting. KF_G3, the single descriptive adverse-condition KF run, is reported without an estimable confidence interval or population-level adverse-illumination inference. A further limitation is that the inertial prediction pathway was inactive in the logged Kalman-filter runs because the inertial coupling coefficient &amp;amp;alpha;=0. The main contribution is a reproducible run-level evaluation of confidence-gated estimator&amp;amp;ndash;controller coupling that distinguishes supported evidence from observational and descriptive cases.</p>
	]]></content:encoded>

	<dc:title>Run-Level Evaluation of a Confidence-Gated Kalman Lane-Keeping Architecture for a 1:10-Scale Autonomous Vehicle</dc:title>
			<dc:creator>Rafael Reveles-Martínez</dc:creator>
			<dc:creator>Hamurabi Gamboa-Rosales</dc:creator>
			<dc:creator>Huizilopoztli Luna-García</dc:creator>
			<dc:creator>Erika Sánchez-Femat</dc:creator>
			<dc:creator>Javier Saldívar-Pérez</dc:creator>
			<dc:creator>Flabio D. Mirelez-Delgado</dc:creator>
			<dc:creator>Umanel A. Hernández-González</dc:creator>
			<dc:creator>Carlos E. Galván-Tejada</dc:creator>
			<dc:creator>Jorge I. Galván-Tejada</dc:creator>
			<dc:creator>José M. Celaya-Padilla</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040129</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-08-13</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-08-13</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>129</prism:startingPage>
		<prism:doi>10.3390/automation7040129</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/129</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/128">

	<title>Automation, Vol. 7, Pages 128: Multi-Objective Optimized Fuzzy Logic Control for Robust Automated Insulin Infusion in Type I Diabetes</title>
	<link>https://www.mdpi.com/2673-4052/7/4/128</link>
	<description>Type I diabetes mellitus (T1DM) is a chronic metabolic disease resulting from insufficient insulin secretion into the bloodstream&amp;amp;sbquo; causing elevated blood glucose concentrations to dangerous levels&amp;amp;#8228; Automated regulation of blood glucose levels in T1DM can be modeled as a nonlinear&amp;amp;sbquo; uncertain&amp;amp;sbquo; and disturbance-affected closed-loop control process with a time delay&amp;amp;sbquo; time-varying insulin sensitivity, and imperfect glucose measurements. This paper presents the design, multi-objective tuning, and robustness evaluation of a fuzzy logic controller (FLC) for automated insulin-infusion regulation. The proposed FLC uses the glucose tracking error and its time derivative as feedback signals to determine the required insulin control action and maintain glucose within the desired range of (70&amp;amp;ndash;160 mg/dL). The controller parameters are optimized using the Non-Dominated Sorting Genetic Algorithm II (NSGA-II) to address three competing control objectives: minimizing hypoglycemia risk, minimizing hyperglycemia risk, and reducing total insulin usage. The resulting Pareto-optimal solutions provide a set of trade-off controller designs for decision-makers based on safety, performance, and insulin-efficiency requirements. The robustness of the proposed automated control framework is evaluated under challenging operating conditions, including elevated initial glucose levels, model-parameter uncertainties, external disturbances, variations in insulin sensitivity, distorted glucose measurements, and delayed insulin infusion. A comparative study with a linear quadratic regulator-based controller (LQRC) is conducted as a benchmark. Simulation results demonstrate that the optimized FLC provides superior closed-loop performance and stronger robustness than the LQRC across all tested scenarios. The proposed fuzzy-control framework, therefore, offers a promising automation-based strategy for resilient glucose regulation under uncertainty, measurement imperfections, and actuation delays.</description>
	<pubDate>2026-08-08</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 128: Multi-Objective Optimized Fuzzy Logic Control for Robust Automated Insulin Infusion in Type I Diabetes</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/128">doi: 10.3390/automation7040128</a></p>
	<p>Authors:
		Raya Abu Shaker
		Yousef Sardahi
		Ahmad Alshorman
		</p>
	<p>Type I diabetes mellitus (T1DM) is a chronic metabolic disease resulting from insufficient insulin secretion into the bloodstream&amp;amp;sbquo; causing elevated blood glucose concentrations to dangerous levels&amp;amp;#8228; Automated regulation of blood glucose levels in T1DM can be modeled as a nonlinear&amp;amp;sbquo; uncertain&amp;amp;sbquo; and disturbance-affected closed-loop control process with a time delay&amp;amp;sbquo; time-varying insulin sensitivity, and imperfect glucose measurements. This paper presents the design, multi-objective tuning, and robustness evaluation of a fuzzy logic controller (FLC) for automated insulin-infusion regulation. The proposed FLC uses the glucose tracking error and its time derivative as feedback signals to determine the required insulin control action and maintain glucose within the desired range of (70&amp;amp;ndash;160 mg/dL). The controller parameters are optimized using the Non-Dominated Sorting Genetic Algorithm II (NSGA-II) to address three competing control objectives: minimizing hypoglycemia risk, minimizing hyperglycemia risk, and reducing total insulin usage. The resulting Pareto-optimal solutions provide a set of trade-off controller designs for decision-makers based on safety, performance, and insulin-efficiency requirements. The robustness of the proposed automated control framework is evaluated under challenging operating conditions, including elevated initial glucose levels, model-parameter uncertainties, external disturbances, variations in insulin sensitivity, distorted glucose measurements, and delayed insulin infusion. A comparative study with a linear quadratic regulator-based controller (LQRC) is conducted as a benchmark. Simulation results demonstrate that the optimized FLC provides superior closed-loop performance and stronger robustness than the LQRC across all tested scenarios. The proposed fuzzy-control framework, therefore, offers a promising automation-based strategy for resilient glucose regulation under uncertainty, measurement imperfections, and actuation delays.</p>
	]]></content:encoded>

	<dc:title>Multi-Objective Optimized Fuzzy Logic Control for Robust Automated Insulin Infusion in Type I Diabetes</dc:title>
			<dc:creator>Raya Abu Shaker</dc:creator>
			<dc:creator>Yousef Sardahi</dc:creator>
			<dc:creator>Ahmad Alshorman</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040128</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-08-08</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-08-08</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>128</prism:startingPage>
		<prism:doi>10.3390/automation7040128</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/128</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/127">

	<title>Automation, Vol. 7, Pages 127: Ranking Inversion in Risk-Parameterised UAV Path Planning for Wildfire Emergency Response</title>
	<link>https://www.mdpi.com/2673-4052/7/4/127</link>
	<description>Wildfires generate rapidly evolving hazard landscapes that disrupt ground-based logistics and render conventional disaster-response operations ineffective, motivating the use of unmanned aerial vehicles (UAVs) in civil-protection missions such as medical resupply, casualty search-and-rescue, and perimeter surveillance. Existing evaluations, however, share a common limitation: they assess performance using navigation-centric metrics—primarily success rate—without accounting for the temporal value of the mission objective. This paper characterises, within each planner family, how a single risk coefficient ρ governs the trade-off between navigation success and time-decaying mission value: holding each family’s algorithm and replan trigger fixed and sweeping only ρ isolates its effect, so that the risk setting maximising a family’s navigation success need not maximise its mission value. To study this, FLARE is introduced, a deterministic benchmark for fire-landscape adaptive risk evaluation, inspired by recent Greek wildfire events; all hazard dynamics (fire spread, structural collapse, moving obstacles, dynamic no-fly zones) are modelled as benchmark abstractions rather than incident-specific reconstructions. FLARE evaluates eight planners spanning six risk-handling paradigm families, isolating the risk coefficient by sweeping ρ within each family (algorithm fixed) with A* as the risk-blind static reference, and quantifies mission impact—medication efficacy, casualty survival, and data freshness—through a strictly time-decreasing mission-score function. The mission-value leverage of ρ is strongly family-dependent: decisive for the incremental soft-cost-inflation family—whose success-optimal ρ collapses its mission value (0.91 → 0.41)—strong for the worst-case (CVaR) family, moderate for the sampling family, and negligible for the reactive, hard-threshold and frequent-replan families. In some families the success-optimal and mission-optimal ρ diverge—a within-family ranking inversion—while in others they coincide; because the algorithm is fixed across each sweep, the divergence is attributable to ρ alone. Success of navigation is therefore a necessary but insufficient proxy for mission effectiveness, and the risk coefficient is a meaningful per-family parameter for mission-aware configuration.</description>
	<pubDate>2026-08-07</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 127: Ranking Inversion in Risk-Parameterised UAV Path Planning for Wildfire Emergency Response</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/127">doi: 10.3390/automation7040127</a></p>
	<p>Authors:
		Konstantinos Zervakis
		Ilias Panagiotopoulos
		</p>
	<p>Wildfires generate rapidly evolving hazard landscapes that disrupt ground-based logistics and render conventional disaster-response operations ineffective, motivating the use of unmanned aerial vehicles (UAVs) in civil-protection missions such as medical resupply, casualty search-and-rescue, and perimeter surveillance. Existing evaluations, however, share a common limitation: they assess performance using navigation-centric metrics—primarily success rate—without accounting for the temporal value of the mission objective. This paper characterises, within each planner family, how a single risk coefficient ρ governs the trade-off between navigation success and time-decaying mission value: holding each family’s algorithm and replan trigger fixed and sweeping only ρ isolates its effect, so that the risk setting maximising a family’s navigation success need not maximise its mission value. To study this, FLARE is introduced, a deterministic benchmark for fire-landscape adaptive risk evaluation, inspired by recent Greek wildfire events; all hazard dynamics (fire spread, structural collapse, moving obstacles, dynamic no-fly zones) are modelled as benchmark abstractions rather than incident-specific reconstructions. FLARE evaluates eight planners spanning six risk-handling paradigm families, isolating the risk coefficient by sweeping ρ within each family (algorithm fixed) with A* as the risk-blind static reference, and quantifies mission impact—medication efficacy, casualty survival, and data freshness—through a strictly time-decreasing mission-score function. The mission-value leverage of ρ is strongly family-dependent: decisive for the incremental soft-cost-inflation family—whose success-optimal ρ collapses its mission value (0.91 → 0.41)—strong for the worst-case (CVaR) family, moderate for the sampling family, and negligible for the reactive, hard-threshold and frequent-replan families. In some families the success-optimal and mission-optimal ρ diverge—a within-family ranking inversion—while in others they coincide; because the algorithm is fixed across each sweep, the divergence is attributable to ρ alone. Success of navigation is therefore a necessary but insufficient proxy for mission effectiveness, and the risk coefficient is a meaningful per-family parameter for mission-aware configuration.</p>
	]]></content:encoded>

	<dc:title>Ranking Inversion in Risk-Parameterised UAV Path Planning for Wildfire Emergency Response</dc:title>
			<dc:creator>Konstantinos Zervakis</dc:creator>
			<dc:creator>Ilias Panagiotopoulos</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040127</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-08-07</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-08-07</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>127</prism:startingPage>
		<prism:doi>10.3390/automation7040127</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/127</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/126">

	<title>Automation, Vol. 7, Pages 126: An Explainable IoT-Enabled Predictive Maintenance Framework Using Digital Twin and Multi-Sensor Machine Learning</title>
	<link>https://www.mdpi.com/2673-4052/7/4/126</link>
	<description>In Industry 4.0 manufacturing, predictive maintenance has emerged as a key focus to ensure the reliability of operations, minimize downtime, and contribute to equipment safety. This study proposes a predictive maintenance framework for additive manufacturing systems (AMSs) using an explainable artificial intelligence (XAI) and machine learning (ML) approach, which is supported by real-time multi-sensor monitoring and is synchronized with a digital twin architecture. A heterogeneous dataset of over 10,000 observations and 13 sensor attributes was gathered from a sensing architecture with ESP32 cameras designed to handle heterogeneous signals from thermal, environmental, mechanical and safety-related sensors. Domain-aware feature engineering was done to gain insights into operation indicators such as temperature instability, vibration degradation, smoke risk, humidity anomalies and aggregated maintenance risk scores. Multiple predictive models, such as the Random Forest model, XGBoost model, Logistic Regression model, K-Nearest Neighbours classifier, Multi-Layer Perceptron model, and ARIMA forecast model, were comparatively assessed under highly imbalanced maintenance conditions. The results showed that ensemble learning methods, especially XGBoost and MLP, had better recall and ROC-AUC for fault detection of maintenance-critical problems. SHAP and LIME analyses then showed that a number of physically meaningful indicators, including thermal instability, vibration anomalies, smoke-related features, and aggregated risk scores, have a significant impact on maintenance predictions and hence provide better operational interpretability and maintenance transparency.</description>
	<pubDate>2026-08-07</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 126: An Explainable IoT-Enabled Predictive Maintenance Framework Using Digital Twin and Multi-Sensor Machine Learning</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/126">doi: 10.3390/automation7040126</a></p>
	<p>Authors:
		Chitranjanjit Kaur
		Sumit Chopra
		Chitta Tripathy
		</p>
	<p>In Industry 4.0 manufacturing, predictive maintenance has emerged as a key focus to ensure the reliability of operations, minimize downtime, and contribute to equipment safety. This study proposes a predictive maintenance framework for additive manufacturing systems (AMSs) using an explainable artificial intelligence (XAI) and machine learning (ML) approach, which is supported by real-time multi-sensor monitoring and is synchronized with a digital twin architecture. A heterogeneous dataset of over 10,000 observations and 13 sensor attributes was gathered from a sensing architecture with ESP32 cameras designed to handle heterogeneous signals from thermal, environmental, mechanical and safety-related sensors. Domain-aware feature engineering was done to gain insights into operation indicators such as temperature instability, vibration degradation, smoke risk, humidity anomalies and aggregated maintenance risk scores. Multiple predictive models, such as the Random Forest model, XGBoost model, Logistic Regression model, K-Nearest Neighbours classifier, Multi-Layer Perceptron model, and ARIMA forecast model, were comparatively assessed under highly imbalanced maintenance conditions. The results showed that ensemble learning methods, especially XGBoost and MLP, had better recall and ROC-AUC for fault detection of maintenance-critical problems. SHAP and LIME analyses then showed that a number of physically meaningful indicators, including thermal instability, vibration anomalies, smoke-related features, and aggregated risk scores, have a significant impact on maintenance predictions and hence provide better operational interpretability and maintenance transparency.</p>
	]]></content:encoded>

	<dc:title>An Explainable IoT-Enabled Predictive Maintenance Framework Using Digital Twin and Multi-Sensor Machine Learning</dc:title>
			<dc:creator>Chitranjanjit Kaur</dc:creator>
			<dc:creator>Sumit Chopra</dc:creator>
			<dc:creator>Chitta Tripathy</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040126</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-08-07</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-08-07</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>126</prism:startingPage>
		<prism:doi>10.3390/automation7040126</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/126</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/125">

	<title>Automation, Vol. 7, Pages 125: Feature-Based Machine Learning Framework for Multi-Source NH3 Dataset Analysis</title>
	<link>https://www.mdpi.com/2673-4052/7/4/125</link>
	<description>Accurate monitoring of atmospheric ammonia (NH3) is important for air-quality assessment and sustainable agriculture, but available datasets differ in spatial resolution, temporal coverage, units, and physical meaning. This study presents a feature-based machine-learning framework with train&amp;amp;ndash;test leakage-controlled preprocessing to evaluate relative NH3 classification consistency across three datasets over China: CAMS GEI, CAMS EAC4, and MEIC. Statistical descriptors were extracted for four spatial&amp;amp;ndash;temporal cases: Province&amp;amp;ndash;Year, Zone&amp;amp;ndash;Year, Province&amp;amp;ndash;Season&amp;amp;ndash;Year, and Zone&amp;amp;ndash;Season&amp;amp;ndash;Year. Six classifiers were evaluated using chronological testing and cross-validation. CAMS GEI provided the broadest spatial&amp;amp;ndash;temporal coverage, while MEIC showed comparatively stable classifier behavior. The best-performing ensemble models commonly achieved accuracies between 0.97 and 0.99 in the main province-level cases. Mean Absolute Value (MAV) was the leading feature in several province-level analyses, with a maximum reported contribution of 64.26%. These scores describe the separability of threshold-derived reference classes and should not be interpreted as an independent physical prediction of NH3 from external atmospheric drivers.</description>
	<pubDate>2026-08-07</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 125: Feature-Based Machine Learning Framework for Multi-Source NH3 Dataset Analysis</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/125">doi: 10.3390/automation7040125</a></p>
	<p>Authors:
		Ata Jahangir Moshayedi
		Babar Hussain Shah
		Amir Sohail Khan
		Seyyed Ali Eftekhari
		Amin Kolahdooz
		David Bassir
		</p>
	<p>Accurate monitoring of atmospheric ammonia (NH3) is important for air-quality assessment and sustainable agriculture, but available datasets differ in spatial resolution, temporal coverage, units, and physical meaning. This study presents a feature-based machine-learning framework with train&amp;amp;ndash;test leakage-controlled preprocessing to evaluate relative NH3 classification consistency across three datasets over China: CAMS GEI, CAMS EAC4, and MEIC. Statistical descriptors were extracted for four spatial&amp;amp;ndash;temporal cases: Province&amp;amp;ndash;Year, Zone&amp;amp;ndash;Year, Province&amp;amp;ndash;Season&amp;amp;ndash;Year, and Zone&amp;amp;ndash;Season&amp;amp;ndash;Year. Six classifiers were evaluated using chronological testing and cross-validation. CAMS GEI provided the broadest spatial&amp;amp;ndash;temporal coverage, while MEIC showed comparatively stable classifier behavior. The best-performing ensemble models commonly achieved accuracies between 0.97 and 0.99 in the main province-level cases. Mean Absolute Value (MAV) was the leading feature in several province-level analyses, with a maximum reported contribution of 64.26%. These scores describe the separability of threshold-derived reference classes and should not be interpreted as an independent physical prediction of NH3 from external atmospheric drivers.</p>
	]]></content:encoded>

	<dc:title>Feature-Based Machine Learning Framework for Multi-Source NH3 Dataset Analysis</dc:title>
			<dc:creator>Ata Jahangir Moshayedi</dc:creator>
			<dc:creator>Babar Hussain Shah</dc:creator>
			<dc:creator>Amir Sohail Khan</dc:creator>
			<dc:creator>Seyyed Ali Eftekhari</dc:creator>
			<dc:creator>Amin Kolahdooz</dc:creator>
			<dc:creator>David Bassir</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040125</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-08-07</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-08-07</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>125</prism:startingPage>
		<prism:doi>10.3390/automation7040125</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/125</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/124">

	<title>Automation, Vol. 7, Pages 124: SMILE: Scalable Modular Instrumentation for Laboratory Experiments</title>
	<link>https://www.mdpi.com/2673-4052/7/4/124</link>
	<description>Scalable Modular Instrumentation for Laboratory Experiments (SMILE) is a lightweight framework for the rapid development and networking of laboratory instrumentation demonstrated with a low-cost Arduino micro-controller. The approach extends the fast-prototyping paradigm of Arduino by enabling a seamless transition from standalone devices to distributed, network-accessible systems without requiring complex control infrastructures. The system architecture follows a simplistic and intuitive development workflow: devices are first implemented and defined through a human-readable serial interface, which is then reused without modification by a Python-based driver. The driver can be directly accessed or enabled as a network service via ZeroRPC, allowing transparent remote access to instrument&amp;amp;rsquo;s functionality. A key feature of SMILE is the automatic mapping of serial commands to Python functions, which facilitates immediate integration of newly defined device commands into higher-level control and automation workflows. SMILE design allows heterogeneous system integration with both custom-built instruments and laboratory equipment with standard interfaces (e.g., GPIB, RS232/485, USB, Ethernet) to be incorporated into a unified distributed system through lightweight software layers. Presented demonstrator examples and test results show that SMILE provides a lightweight and accessible approach for physics laboratory automation, conceptually inspired by distributed control systems such as TANGO and EPICS, while remaining focused on small-scale experiments and rapid prototyping.</description>
	<pubDate>2026-08-04</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 124: SMILE: Scalable Modular Instrumentation for Laboratory Experiments</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/124">doi: 10.3390/automation7040124</a></p>
	<p>Authors:
		Kamen Kamenov
		Viktor Angelov
		Lyubomir Karlov
		Krastyo Buchkov
		</p>
	<p>Scalable Modular Instrumentation for Laboratory Experiments (SMILE) is a lightweight framework for the rapid development and networking of laboratory instrumentation demonstrated with a low-cost Arduino micro-controller. The approach extends the fast-prototyping paradigm of Arduino by enabling a seamless transition from standalone devices to distributed, network-accessible systems without requiring complex control infrastructures. The system architecture follows a simplistic and intuitive development workflow: devices are first implemented and defined through a human-readable serial interface, which is then reused without modification by a Python-based driver. The driver can be directly accessed or enabled as a network service via ZeroRPC, allowing transparent remote access to instrument&amp;amp;rsquo;s functionality. A key feature of SMILE is the automatic mapping of serial commands to Python functions, which facilitates immediate integration of newly defined device commands into higher-level control and automation workflows. SMILE design allows heterogeneous system integration with both custom-built instruments and laboratory equipment with standard interfaces (e.g., GPIB, RS232/485, USB, Ethernet) to be incorporated into a unified distributed system through lightweight software layers. Presented demonstrator examples and test results show that SMILE provides a lightweight and accessible approach for physics laboratory automation, conceptually inspired by distributed control systems such as TANGO and EPICS, while remaining focused on small-scale experiments and rapid prototyping.</p>
	]]></content:encoded>

	<dc:title>SMILE: Scalable Modular Instrumentation for Laboratory Experiments</dc:title>
			<dc:creator>Kamen Kamenov</dc:creator>
			<dc:creator>Viktor Angelov</dc:creator>
			<dc:creator>Lyubomir Karlov</dc:creator>
			<dc:creator>Krastyo Buchkov</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040124</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-08-04</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-08-04</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>124</prism:startingPage>
		<prism:doi>10.3390/automation7040124</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/124</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/123">

	<title>Automation, Vol. 7, Pages 123: A Neural Approach for Position&amp;ndash;Orientation Tracking of the Stewart Platform with Disturbance Suppression</title>
	<link>https://www.mdpi.com/2673-4052/7/4/123</link>
	<description>In this paper, the position&amp;amp;ndash;orientation tracking of the Stewart platform is investigated. We focus on the target-oriented tracking task, which usually occurs in the application of spotlights and cameras. Specifically, the mobile plate of the Stewart platform is always oriented to an immobile point while tracking a desired path. Based on the velocity-level kinematics and zeroing neural dynamics (ZND), a kinematic tracking model is proposed for the position&amp;amp;ndash;orientation tracking task. In addition, a robust ZND (RZND) model is further proposed against two kinds of disturbances. Theoretical analyses are presented to show the convergence properties of the proposed models. Discrete-time algorithms of the models are also developed for the convenient implementation. According to the comparative simulations, both the ZND and RZND algorithms accomplish the position&amp;amp;ndash;orientation tracking task without the disturbance, and the disturbance-suppression capability of the RZND algorithm is substantiated.</description>
	<pubDate>2026-08-03</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 123: A Neural Approach for Position&amp;ndash;Orientation Tracking of the Stewart Platform with Disturbance Suppression</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/123">doi: 10.3390/automation7040123</a></p>
	<p>Authors:
		Yunong Zhang
		Jielong Chen
		Zhuosong Fu
		Guangyu Long
		</p>
	<p>In this paper, the position&amp;amp;ndash;orientation tracking of the Stewart platform is investigated. We focus on the target-oriented tracking task, which usually occurs in the application of spotlights and cameras. Specifically, the mobile plate of the Stewart platform is always oriented to an immobile point while tracking a desired path. Based on the velocity-level kinematics and zeroing neural dynamics (ZND), a kinematic tracking model is proposed for the position&amp;amp;ndash;orientation tracking task. In addition, a robust ZND (RZND) model is further proposed against two kinds of disturbances. Theoretical analyses are presented to show the convergence properties of the proposed models. Discrete-time algorithms of the models are also developed for the convenient implementation. According to the comparative simulations, both the ZND and RZND algorithms accomplish the position&amp;amp;ndash;orientation tracking task without the disturbance, and the disturbance-suppression capability of the RZND algorithm is substantiated.</p>
	]]></content:encoded>

	<dc:title>A Neural Approach for Position&amp;amp;ndash;Orientation Tracking of the Stewart Platform with Disturbance Suppression</dc:title>
			<dc:creator>Yunong Zhang</dc:creator>
			<dc:creator>Jielong Chen</dc:creator>
			<dc:creator>Zhuosong Fu</dc:creator>
			<dc:creator>Guangyu Long</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040123</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-08-03</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-08-03</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>123</prism:startingPage>
		<prism:doi>10.3390/automation7040123</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/123</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/122">

	<title>Automation, Vol. 7, Pages 122: From Optimisation to Closed-Loop Urban Automation: A Conceptual Framework for Spatial Intelligence and Physical AI in AI&amp;ndash;IoT-Enabled Smart Cities</title>
	<link>https://www.mdpi.com/2673-4052/7/4/122</link>
	<description>Artificial intelligence (AI) and the Internet of Things (IoT) are converging to create densely sensed, connected, and increasingly automated urban environments. However, research on AI&amp;amp;ndash;IoT integration remains dominated by optimisation-centric perspectives that under-specify spatial reasoning, physical actuation, and institutional accountability. This perspective addresses that gap by developing closed-loop urban automation as an analytical lens for understanding how AI&amp;amp;ndash;IoT systems move from urban sensing to consequential intervention. Drawing on a narrative, theory-driven synthesis of literature on Urban AI, IoT, Physical AI, spatial intelligence, digital twins, robotics, automation, and governance, the paper differentiates the framework from AIoT, cyber-physical systems, embodied AI, and optimisation-centric smart-city models. It identifies the distinctive conditions of urban automation, proposes six examinable propositions, and outlines implications for planning support, intelligent control, human oversight, and democratic legitimacy.</description>
	<pubDate>2026-08-03</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 122: From Optimisation to Closed-Loop Urban Automation: A Conceptual Framework for Spatial Intelligence and Physical AI in AI&amp;ndash;IoT-Enabled Smart Cities</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/122">doi: 10.3390/automation7040122</a></p>
	<p>Authors:
		Alok Tiwari
		Yasser Qaffas
		</p>
	<p>Artificial intelligence (AI) and the Internet of Things (IoT) are converging to create densely sensed, connected, and increasingly automated urban environments. However, research on AI&amp;amp;ndash;IoT integration remains dominated by optimisation-centric perspectives that under-specify spatial reasoning, physical actuation, and institutional accountability. This perspective addresses that gap by developing closed-loop urban automation as an analytical lens for understanding how AI&amp;amp;ndash;IoT systems move from urban sensing to consequential intervention. Drawing on a narrative, theory-driven synthesis of literature on Urban AI, IoT, Physical AI, spatial intelligence, digital twins, robotics, automation, and governance, the paper differentiates the framework from AIoT, cyber-physical systems, embodied AI, and optimisation-centric smart-city models. It identifies the distinctive conditions of urban automation, proposes six examinable propositions, and outlines implications for planning support, intelligent control, human oversight, and democratic legitimacy.</p>
	]]></content:encoded>

	<dc:title>From Optimisation to Closed-Loop Urban Automation: A Conceptual Framework for Spatial Intelligence and Physical AI in AI&amp;amp;ndash;IoT-Enabled Smart Cities</dc:title>
			<dc:creator>Alok Tiwari</dc:creator>
			<dc:creator>Yasser Qaffas</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040122</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-08-03</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-08-03</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Perspective</prism:section>
	<prism:startingPage>122</prism:startingPage>
		<prism:doi>10.3390/automation7040122</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/122</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/121">

	<title>Automation, Vol. 7, Pages 121: Bridging the Reality Gap in Hyperstatic Mechanisms: Nonlinear Stribeck Friction Modeling and Virtual Certification via SiL Co-Simulation</title>
	<link>https://www.mdpi.com/2673-4052/7/4/121</link>
	<description>In aerospace manufacturing, validating heavy-duty automated production tooling and Ground Support Equipment (GSE) traditionally requires costly and time-consuming physical proof load testing. This study proposes a novel Virtual Certification framework that utilizes a high-fidelity Multiphysical Digital Twin driven by a Software-in-the-Loop (SiL) co-simulation architecture (integrating Siemens NX MCD, SIMIT, and TIA Portal) to retroactively diagnose mechanical failures and virtually validate design modifications prior to physical manufacturing. The dual-focus methodology is rigorously applied to a physical case study: an over-constrained (hyperstatic) 4-point aerospace lifting system designed for a 26.48 kN fuselage section that suffered a catastrophic mechanical stall during a 39.24 kN physical proof load verification. While conventional static dimensioning models erroneously predicted a nominal drive torque of only 4.56 Nm, the high-fidelity dynamic twin (incorporating a non-linear exponential Stribeck friction model) calculated the transient mechanical resistance causing the stall, capturing a peak load of 46.2 Nm at the motor shaft. The SiL co-simulation revealed that the rigid positional synchronization logic enforced by the PLC inadvertently amplified localized boundary friction, driving the actuators beyond their rated 6.4 Nm capacity. Based on this forensic diagnosis, a remedial powertrain featuring an 8.0 Nm stepper motor coupled with a 16:1 planetary gearbox was integrated and virtually certified. The framework confirmed that the upgraded architecture successfully attenuated the hyperstatic resistance, reflecting a peak load of only 3.0 Nm at the motor shaft and guaranteeing a stable Safety Factor of 2.66. By bridging the reality gap without iterative physical prototyping, this framework establishes a scalable, &amp;amp;ldquo;First-Time-Right&amp;amp;rdquo; validation paradigm for multi-point automated manufacturing mechanisms.</description>
	<pubDate>2026-08-01</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 121: Bridging the Reality Gap in Hyperstatic Mechanisms: Nonlinear Stribeck Friction Modeling and Virtual Certification via SiL Co-Simulation</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/121">doi: 10.3390/automation7040121</a></p>
	<p>Authors:
		Yakup Kılıçaslan
		Sami Karadeniz
		</p>
	<p>In aerospace manufacturing, validating heavy-duty automated production tooling and Ground Support Equipment (GSE) traditionally requires costly and time-consuming physical proof load testing. This study proposes a novel Virtual Certification framework that utilizes a high-fidelity Multiphysical Digital Twin driven by a Software-in-the-Loop (SiL) co-simulation architecture (integrating Siemens NX MCD, SIMIT, and TIA Portal) to retroactively diagnose mechanical failures and virtually validate design modifications prior to physical manufacturing. The dual-focus methodology is rigorously applied to a physical case study: an over-constrained (hyperstatic) 4-point aerospace lifting system designed for a 26.48 kN fuselage section that suffered a catastrophic mechanical stall during a 39.24 kN physical proof load verification. While conventional static dimensioning models erroneously predicted a nominal drive torque of only 4.56 Nm, the high-fidelity dynamic twin (incorporating a non-linear exponential Stribeck friction model) calculated the transient mechanical resistance causing the stall, capturing a peak load of 46.2 Nm at the motor shaft. The SiL co-simulation revealed that the rigid positional synchronization logic enforced by the PLC inadvertently amplified localized boundary friction, driving the actuators beyond their rated 6.4 Nm capacity. Based on this forensic diagnosis, a remedial powertrain featuring an 8.0 Nm stepper motor coupled with a 16:1 planetary gearbox was integrated and virtually certified. The framework confirmed that the upgraded architecture successfully attenuated the hyperstatic resistance, reflecting a peak load of only 3.0 Nm at the motor shaft and guaranteeing a stable Safety Factor of 2.66. By bridging the reality gap without iterative physical prototyping, this framework establishes a scalable, &amp;amp;ldquo;First-Time-Right&amp;amp;rdquo; validation paradigm for multi-point automated manufacturing mechanisms.</p>
	]]></content:encoded>

	<dc:title>Bridging the Reality Gap in Hyperstatic Mechanisms: Nonlinear Stribeck Friction Modeling and Virtual Certification via SiL Co-Simulation</dc:title>
			<dc:creator>Yakup Kılıçaslan</dc:creator>
			<dc:creator>Sami Karadeniz</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040121</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-08-01</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-08-01</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>121</prism:startingPage>
		<prism:doi>10.3390/automation7040121</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/121</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/120">

	<title>Automation, Vol. 7, Pages 120: A Lightweight Deep Learning Framework for Real-Time Brinjal Detection Under Field Conditions</title>
	<link>https://www.mdpi.com/2673-4052/7/4/120</link>
	<description>Brinjal (Solanum melongena L.) is an important vegetable crop worldwide, but its cultivation faces challenges from pests, diseases, and variable environmental conditions that negatively affect quality and yield. Accurate fruit detection in natural field conditions is essential for yield estimation and perception module of automated harvesting, but existing deep learning techniques often require substantial computational support, restricting their deployment on edge devices. This study addresses this gap by evaluating the Faster Objects, More Objects (FOMO) model&amp;amp;mdash;a lightweight architecture for resource-constrained platforms&amp;amp;mdash;for brinjal detection under diverse field conditions. A dataset of 1500 images was captured under varying illumination and growth stages, annotated using a bounding box-based approach, and used to train FOMO models with 25, 50, and 100 epochs via transfer learning. Post-training quantization to INT8 format was applied to assess improvements in computational efficiency. The Float32 model achieved a precision of 0.827, recall of 0.915, and F1-score of 0.869 at 100 epochs. The INT8-quantized model maintained comparable accuracy (precision 0.829, recall 0.908, F1-score 0.866) while reducing model size by 63.33% (from 0.30 MB to 0.11 MB), inference time by 62.02% (from 65.3 ms to 24.8 ms), and RAM usage by 73.02% (from 887.2 KB to 239.4 KB). These results demonstrate that FOMO combined with INT8 quantization provides an efficient, accurate solution for real-time brinjal detection on edge platforms, supporting the advancement of precision agriculture through intelligent crop monitoring and serving as a perception module for future robotic harvesting systems.</description>
	<pubDate>2026-08-01</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 120: A Lightweight Deep Learning Framework for Real-Time Brinjal Detection Under Field Conditions</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/120">doi: 10.3390/automation7040120</a></p>
	<p>Authors:
		Abhishek Pandey
		Pramod Kumar Sahoo
		Tapan Kumar Khura
		Dilip Kumar Kushwaha
		Roaf Ahmad Parray
		Jeetendra Kumar Ranjan
		Md. Ashraful Haque
		Susheel Kumar Sarkar
		Rohit Gaddamwar
		Nrusingh Charan Pradhan
		</p>
	<p>Brinjal (Solanum melongena L.) is an important vegetable crop worldwide, but its cultivation faces challenges from pests, diseases, and variable environmental conditions that negatively affect quality and yield. Accurate fruit detection in natural field conditions is essential for yield estimation and perception module of automated harvesting, but existing deep learning techniques often require substantial computational support, restricting their deployment on edge devices. This study addresses this gap by evaluating the Faster Objects, More Objects (FOMO) model&amp;amp;mdash;a lightweight architecture for resource-constrained platforms&amp;amp;mdash;for brinjal detection under diverse field conditions. A dataset of 1500 images was captured under varying illumination and growth stages, annotated using a bounding box-based approach, and used to train FOMO models with 25, 50, and 100 epochs via transfer learning. Post-training quantization to INT8 format was applied to assess improvements in computational efficiency. The Float32 model achieved a precision of 0.827, recall of 0.915, and F1-score of 0.869 at 100 epochs. The INT8-quantized model maintained comparable accuracy (precision 0.829, recall 0.908, F1-score 0.866) while reducing model size by 63.33% (from 0.30 MB to 0.11 MB), inference time by 62.02% (from 65.3 ms to 24.8 ms), and RAM usage by 73.02% (from 887.2 KB to 239.4 KB). These results demonstrate that FOMO combined with INT8 quantization provides an efficient, accurate solution for real-time brinjal detection on edge platforms, supporting the advancement of precision agriculture through intelligent crop monitoring and serving as a perception module for future robotic harvesting systems.</p>
	]]></content:encoded>

	<dc:title>A Lightweight Deep Learning Framework for Real-Time Brinjal Detection Under Field Conditions</dc:title>
			<dc:creator>Abhishek Pandey</dc:creator>
			<dc:creator>Pramod Kumar Sahoo</dc:creator>
			<dc:creator>Tapan Kumar Khura</dc:creator>
			<dc:creator>Dilip Kumar Kushwaha</dc:creator>
			<dc:creator>Roaf Ahmad Parray</dc:creator>
			<dc:creator>Jeetendra Kumar Ranjan</dc:creator>
			<dc:creator>Md. Ashraful Haque</dc:creator>
			<dc:creator>Susheel Kumar Sarkar</dc:creator>
			<dc:creator>Rohit Gaddamwar</dc:creator>
			<dc:creator>Nrusingh Charan Pradhan</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040120</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-08-01</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-08-01</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>120</prism:startingPage>
		<prism:doi>10.3390/automation7040120</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/120</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/119">

	<title>Automation, Vol. 7, Pages 119: Reliability-Aware Occupancy Map Merging in Dynamic Environments Using Temporal and Probabilistic Maps</title>
	<link>https://www.mdpi.com/2673-4052/7/4/119</link>
	<description>Occupancy map merging in dynamic environments is challenging because moving objects introduce time-varying disturbances that degrade registration accuracy and structural consistency. This paper proposes a reliability-aware map-merging method that explicitly models temporal validity and probabilistic reliability in occupancy grid maps. The method constructs two complementary maps: a trajectory-guided temporal reliability map that reflects the recency and persistence of cell observations and a probabilistic reliability map that refines detected candidate regions based on temporal persistence and spatial reliability criteria. By suppressing transient clutter before registration, the proposed approach focuses alignment on structurally stable regions and improves merging robustness. Experiments in simulation and real-world indoor environments demonstrate clear improvements over a conventional feature-based baseline, substantially reducing both translation and rotation errors. These results show that incorporating temporal validity and probabilistic reliability can improve occupancy map merging under dynamic conditions.</description>
	<pubDate>2026-08-01</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 119: Reliability-Aware Occupancy Map Merging in Dynamic Environments Using Temporal and Probabilistic Maps</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/119">doi: 10.3390/automation7040119</a></p>
	<p>Authors:
		Hanngyoo Kim
		Seunghwan Lee
		</p>
	<p>Occupancy map merging in dynamic environments is challenging because moving objects introduce time-varying disturbances that degrade registration accuracy and structural consistency. This paper proposes a reliability-aware map-merging method that explicitly models temporal validity and probabilistic reliability in occupancy grid maps. The method constructs two complementary maps: a trajectory-guided temporal reliability map that reflects the recency and persistence of cell observations and a probabilistic reliability map that refines detected candidate regions based on temporal persistence and spatial reliability criteria. By suppressing transient clutter before registration, the proposed approach focuses alignment on structurally stable regions and improves merging robustness. Experiments in simulation and real-world indoor environments demonstrate clear improvements over a conventional feature-based baseline, substantially reducing both translation and rotation errors. These results show that incorporating temporal validity and probabilistic reliability can improve occupancy map merging under dynamic conditions.</p>
	]]></content:encoded>

	<dc:title>Reliability-Aware Occupancy Map Merging in Dynamic Environments Using Temporal and Probabilistic Maps</dc:title>
			<dc:creator>Hanngyoo Kim</dc:creator>
			<dc:creator>Seunghwan Lee</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040119</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-08-01</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-08-01</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>119</prism:startingPage>
		<prism:doi>10.3390/automation7040119</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/119</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/118">

	<title>Automation, Vol. 7, Pages 118: Explainable Transfer Learning for Multi-Crop Plant Disease Identification Under Real-Field Conditions</title>
	<link>https://www.mdpi.com/2673-4052/7/4/118</link>
	<description>Plant diseases have long been considered a major threat to global food production systems. Therefore, early diagnosis is vital to mitigate the risk of these diseases. This task can be challenging, as the number of harmful diseases is substantial. One technology that has gained widespread interest is artificial intelligence, specifically deep learning, which is used to identify plant diseases using leaf patterns. This paper presents two deep learning models, a custom CNN model and a transfer learning model based on the DenseNet-121 architecture. Experiments were carried out using the PlantCity dataset, which consists of twelve subsets of diverse crop species representing fruits, vegetables, and grains with variations among the subsets, including the number of classes, the subset sizes, class distribution, and visual complexity of disease symptoms. The two models were evaluated using multiple metrics, including accuracy, loss, precision, recall, and F1-score. Explainable AI using the LIME technique was deployed to better interpret the acquired results. Results showed that the developed transfer learning model based on DenseNet-121 had superior performance over the CNN model, with accuracies ranging from 86% to 99% across eleven experimented crops. In order to perform an independent experimental validation for the developed model, future work will focus on constructing a local crop dataset captured from Iraqi fields to evaluate the developed models based on the local environment.</description>
	<pubDate>2026-07-28</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 118: Explainable Transfer Learning for Multi-Crop Plant Disease Identification Under Real-Field Conditions</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/118">doi: 10.3390/automation7040118</a></p>
	<p>Authors:
		Hassan A. Jeiad
		Sama S. Samaan
		Omar Janeh
		Saja D. Khudhur
		Amjad J. Humaidi
		</p>
	<p>Plant diseases have long been considered a major threat to global food production systems. Therefore, early diagnosis is vital to mitigate the risk of these diseases. This task can be challenging, as the number of harmful diseases is substantial. One technology that has gained widespread interest is artificial intelligence, specifically deep learning, which is used to identify plant diseases using leaf patterns. This paper presents two deep learning models, a custom CNN model and a transfer learning model based on the DenseNet-121 architecture. Experiments were carried out using the PlantCity dataset, which consists of twelve subsets of diverse crop species representing fruits, vegetables, and grains with variations among the subsets, including the number of classes, the subset sizes, class distribution, and visual complexity of disease symptoms. The two models were evaluated using multiple metrics, including accuracy, loss, precision, recall, and F1-score. Explainable AI using the LIME technique was deployed to better interpret the acquired results. Results showed that the developed transfer learning model based on DenseNet-121 had superior performance over the CNN model, with accuracies ranging from 86% to 99% across eleven experimented crops. In order to perform an independent experimental validation for the developed model, future work will focus on constructing a local crop dataset captured from Iraqi fields to evaluate the developed models based on the local environment.</p>
	]]></content:encoded>

	<dc:title>Explainable Transfer Learning for Multi-Crop Plant Disease Identification Under Real-Field Conditions</dc:title>
			<dc:creator>Hassan A. Jeiad</dc:creator>
			<dc:creator>Sama S. Samaan</dc:creator>
			<dc:creator>Omar Janeh</dc:creator>
			<dc:creator>Saja D. Khudhur</dc:creator>
			<dc:creator>Amjad J. Humaidi</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040118</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-07-28</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-07-28</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>118</prism:startingPage>
		<prism:doi>10.3390/automation7040118</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/118</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/117">

	<title>Automation, Vol. 7, Pages 117: A Hybrid Physics&amp;ndash;AI Framework for Real-Time Emission Monitoring in IIoT-Enabled Industrial Systems</title>
	<link>https://www.mdpi.com/2673-4052/7/4/117</link>
	<description>This paper presents a hybrid physics&amp;amp;ndash;AI framework for real-time emission monitoring in industrial boiler systems within an IIoT-enabled Industry 4.0 environment. The proposed framework integrates physics-based emission estimation with AI-based anomaly detection within a unified operational technology and information technology (OT&amp;amp;ndash;IT) architecture to support continuous environmental monitoring. Process data, including fuel oil consumption, oxygen concentration, temperature, and pressure, are acquired from an industrial boiler through a Siemens programmable logic controller (PLC) using an Open Platform Communications Unified Architecture (OPC UA) communication layer. The acquired measurements are processed at the edge analytics level to estimate the emission rates of carbon monoxide (CO), sulfur dioxide (SO2), nitrogen dioxide (NO2), and particulate matter (PM) using stoichiometric combustion models based on fuel composition and flue gas characteristics. An autoencoder-based anomaly detection model is employed to identify abnormal operating conditions by monitoring the reconstruction error against a predefined threshold. The framework is validated using a PLC-based quasi-real-time prototype that replays one year of historical industrial boiler operating data. The emission estimation results show close agreement with reference engineering calculations, with relative errors below 0.1% across the evaluated operating conditions. The anomaly detection model achieved an F1-score of 96.14% and an AUC of 0.981. An edge monitoring dashboard provides real-time visualization of process variables, estimated emissions, and alarm status, while cloud connectivity supports remote monitoring and long-term data analytics. Overall, the proposed framework demonstrates how existing industrial process data can be utilized to transform conventional offline emission estimation into a continuous OT&amp;amp;ndash;IT monitoring service for legacy industrial environments.</description>
	<pubDate>2026-07-28</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 117: A Hybrid Physics&amp;ndash;AI Framework for Real-Time Emission Monitoring in IIoT-Enabled Industrial Systems</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/117">doi: 10.3390/automation7040117</a></p>
	<p>Authors:
		Abdullah S. Hamoud
		Mahmood Farhan Mosleh
		Salah Al-Zubaidi
		Ramiz M. Shubbar
		</p>
	<p>This paper presents a hybrid physics&amp;amp;ndash;AI framework for real-time emission monitoring in industrial boiler systems within an IIoT-enabled Industry 4.0 environment. The proposed framework integrates physics-based emission estimation with AI-based anomaly detection within a unified operational technology and information technology (OT&amp;amp;ndash;IT) architecture to support continuous environmental monitoring. Process data, including fuel oil consumption, oxygen concentration, temperature, and pressure, are acquired from an industrial boiler through a Siemens programmable logic controller (PLC) using an Open Platform Communications Unified Architecture (OPC UA) communication layer. The acquired measurements are processed at the edge analytics level to estimate the emission rates of carbon monoxide (CO), sulfur dioxide (SO2), nitrogen dioxide (NO2), and particulate matter (PM) using stoichiometric combustion models based on fuel composition and flue gas characteristics. An autoencoder-based anomaly detection model is employed to identify abnormal operating conditions by monitoring the reconstruction error against a predefined threshold. The framework is validated using a PLC-based quasi-real-time prototype that replays one year of historical industrial boiler operating data. The emission estimation results show close agreement with reference engineering calculations, with relative errors below 0.1% across the evaluated operating conditions. The anomaly detection model achieved an F1-score of 96.14% and an AUC of 0.981. An edge monitoring dashboard provides real-time visualization of process variables, estimated emissions, and alarm status, while cloud connectivity supports remote monitoring and long-term data analytics. Overall, the proposed framework demonstrates how existing industrial process data can be utilized to transform conventional offline emission estimation into a continuous OT&amp;amp;ndash;IT monitoring service for legacy industrial environments.</p>
	]]></content:encoded>

	<dc:title>A Hybrid Physics&amp;amp;ndash;AI Framework for Real-Time Emission Monitoring in IIoT-Enabled Industrial Systems</dc:title>
			<dc:creator>Abdullah S. Hamoud</dc:creator>
			<dc:creator>Mahmood Farhan Mosleh</dc:creator>
			<dc:creator>Salah Al-Zubaidi</dc:creator>
			<dc:creator>Ramiz M. Shubbar</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040117</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-07-28</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-07-28</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>117</prism:startingPage>
		<prism:doi>10.3390/automation7040117</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/117</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/116">

	<title>Automation, Vol. 7, Pages 116: Rule-Based Real-Time Energy Management System for Curative Congestion Management in Low-Voltage Distribution Grids</title>
	<link>https://www.mdpi.com/2673-4052/7/4/116</link>
	<description>The electrification of residential demand through electric vehicles (EVs), heat pumps (HPs), photovoltaic (PV) systems, and battery energy storage systems (BESSs) creates new congestion challenges in low-voltage (LV) grids. This study evaluates a transparent, deterministic, and real-time-capable rule-based energy management system (EMS) for curative thermal congestion management within a &amp;amp;sect;14a EnWG-oriented setting. The EMS is implemented in MATLAB/Simulink and tested on a representative four-feeder LV network supplying 56 households. Congestion is detected from maximum phase root-mean-square currents using conservative transformer and feeder thresholds. After a threshold is reached, the EMS first activates available BESS support and then applies simultaneous feeder-wide EV limitation, batched round-robin curtailment, or staged feeder-wide reduction toward 4.2 kW. In the uncontrolled case, the Feeder 3 and transformer overload areas are 62.84 Ah and 48.50 Ah, respectively. All controlled scenarios remove at least 98.70% of the feeder overload and eliminate the transformer overload within the reported numerical precision. The batched strategy requires 328.54 Ah of cumulative feeder-current reduction, compared with 977.34 Ah for simultaneous control and 816.00 Ah for staged control, and achieves the highest feeder-relief efficiency. It therefore provides a balanced trade-off between congestion relief and intervention intensity for the investigated deterministic case.</description>
	<pubDate>2026-07-27</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 116: Rule-Based Real-Time Energy Management System for Curative Congestion Management in Low-Voltage Distribution Grids</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/116">doi: 10.3390/automation7040116</a></p>
	<p>Authors:
		Sajjad Karami
		Payam Teimourzadeh Baboli
		Christian Becker
		</p>
	<p>The electrification of residential demand through electric vehicles (EVs), heat pumps (HPs), photovoltaic (PV) systems, and battery energy storage systems (BESSs) creates new congestion challenges in low-voltage (LV) grids. This study evaluates a transparent, deterministic, and real-time-capable rule-based energy management system (EMS) for curative thermal congestion management within a &amp;amp;sect;14a EnWG-oriented setting. The EMS is implemented in MATLAB/Simulink and tested on a representative four-feeder LV network supplying 56 households. Congestion is detected from maximum phase root-mean-square currents using conservative transformer and feeder thresholds. After a threshold is reached, the EMS first activates available BESS support and then applies simultaneous feeder-wide EV limitation, batched round-robin curtailment, or staged feeder-wide reduction toward 4.2 kW. In the uncontrolled case, the Feeder 3 and transformer overload areas are 62.84 Ah and 48.50 Ah, respectively. All controlled scenarios remove at least 98.70% of the feeder overload and eliminate the transformer overload within the reported numerical precision. The batched strategy requires 328.54 Ah of cumulative feeder-current reduction, compared with 977.34 Ah for simultaneous control and 816.00 Ah for staged control, and achieves the highest feeder-relief efficiency. It therefore provides a balanced trade-off between congestion relief and intervention intensity for the investigated deterministic case.</p>
	]]></content:encoded>

	<dc:title>Rule-Based Real-Time Energy Management System for Curative Congestion Management in Low-Voltage Distribution Grids</dc:title>
			<dc:creator>Sajjad Karami</dc:creator>
			<dc:creator>Payam Teimourzadeh Baboli</dc:creator>
			<dc:creator>Christian Becker</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040116</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-07-27</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-07-27</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>116</prism:startingPage>
		<prism:doi>10.3390/automation7040116</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/116</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/115">

	<title>Automation, Vol. 7, Pages 115: Sustainable and Intelligent Automation Framework for Aerospace Alloys Using Multi-Agent Deep Reinforcement Learning</title>
	<link>https://www.mdpi.com/2673-4052/7/4/115</link>
	<description>Advanced aerospace alloys such as titanium alloy (Ti-6Al-4V) are widely employed in critical applications owing to their excellent strength-to-weight ratio and corrosion resistance. However, machining these alloys remains challenging due to significant tool wear, poor material removal rates, and surface integrity concerns. This research offers a sustainable and intelligent machining framework for a 3 mm thick Ti-6Al-4V alloy, utilizing coated wire electrical discharge machining (WEDM) linked with Multi-Agent Deep Reinforcement Learning (MADRL). A Box&amp;amp;ndash;Behnken experimental design was adopted to gather baseline data for pulse-on time, pulse-off time, servo voltage, and peak current. The MADRL architecture combines cooperative agents to improve MRR, surface roughness, and kerf width concurrently. Beyond performance increase, the sustainability parameters of energy consumption, dielectric fluid use, and wire consumption were also studied. The proposed MADRL significantly improved the material removal rate (MRR) from 1.00 to 1.28 mm3/min and reduced the average surface roughness (Ra) from 1.95 to 1.66 &amp;amp;micro;m, the kerf width from 0.27 to 0.24 mm, and the energy consumption from 122 to 107 J. The results show the potential of the proposed framework for adaptive, sustainable, and high-performance aerospace manufacturing.</description>
	<pubDate>2026-07-24</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 115: Sustainable and Intelligent Automation Framework for Aerospace Alloys Using Multi-Agent Deep Reinforcement Learning</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/115">doi: 10.3390/automation7040115</a></p>
	<p>Authors:
		Nagadeepan Anbazhagan
		Senthilkumar Vagheesan
		K. K. Ilavenil
		</p>
	<p>Advanced aerospace alloys such as titanium alloy (Ti-6Al-4V) are widely employed in critical applications owing to their excellent strength-to-weight ratio and corrosion resistance. However, machining these alloys remains challenging due to significant tool wear, poor material removal rates, and surface integrity concerns. This research offers a sustainable and intelligent machining framework for a 3 mm thick Ti-6Al-4V alloy, utilizing coated wire electrical discharge machining (WEDM) linked with Multi-Agent Deep Reinforcement Learning (MADRL). A Box&amp;amp;ndash;Behnken experimental design was adopted to gather baseline data for pulse-on time, pulse-off time, servo voltage, and peak current. The MADRL architecture combines cooperative agents to improve MRR, surface roughness, and kerf width concurrently. Beyond performance increase, the sustainability parameters of energy consumption, dielectric fluid use, and wire consumption were also studied. The proposed MADRL significantly improved the material removal rate (MRR) from 1.00 to 1.28 mm3/min and reduced the average surface roughness (Ra) from 1.95 to 1.66 &amp;amp;micro;m, the kerf width from 0.27 to 0.24 mm, and the energy consumption from 122 to 107 J. The results show the potential of the proposed framework for adaptive, sustainable, and high-performance aerospace manufacturing.</p>
	]]></content:encoded>

	<dc:title>Sustainable and Intelligent Automation Framework for Aerospace Alloys Using Multi-Agent Deep Reinforcement Learning</dc:title>
			<dc:creator>Nagadeepan Anbazhagan</dc:creator>
			<dc:creator>Senthilkumar Vagheesan</dc:creator>
			<dc:creator>K. K. Ilavenil</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040115</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-07-24</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-07-24</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>115</prism:startingPage>
		<prism:doi>10.3390/automation7040115</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/115</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/114">

	<title>Automation, Vol. 7, Pages 114: Adaptive Fuzzy Feedforward Compensation for High-Precision X&amp;ndash;Y Positioning Systems Driven by Stepper Motors</title>
	<link>https://www.mdpi.com/2673-4052/7/4/114</link>
	<description>High-precision X&amp;amp;ndash;Y positioning systems driven by stepper motors are widely used in industrial automation, manufacturing, and scientific instrumentation. However, fixed feedforward&amp;amp;ndash;feedback controllers may degrade when operating conditions vary, particularly as step frequency changes and the risk of synchronism loss increases. This work proposes an adaptive fuzzy feedforward&amp;amp;ndash;feedback controller for stepper-motor-driven X&amp;amp;ndash;Y positioning systems. The controller uses a Takagi&amp;amp;ndash;Sugeno (T&amp;amp;ndash;S) fuzzy inference system to adjust the proportional, derivative, and feedforward actions according to the tracking error, step frequency, and an auxiliary error-based adaptation variable. The control law is integrated with the inverse kinematics of the platform to generate synchronized step-domain commands, and a practical synchronism-preservation condition is established. Experimental validation on a NEMA 17-based X&amp;amp;ndash;Y platform showed accurate trajectory tracking, with a steady-state error of approximately 1.6[&amp;amp;mu;m] for a trapezoidal profile. For a multi-segment trajectory, the RMSE was 0.0749[mm] without load and 0.0760[mm] under a 7.5[kg] external load. Compared with a conventional PID controller, the proposed method reduced the RMSE from 0.1741[mm] to 0.0749[mm], while preserving motor synchronism.</description>
	<pubDate>2026-07-23</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 114: Adaptive Fuzzy Feedforward Compensation for High-Precision X&amp;ndash;Y Positioning Systems Driven by Stepper Motors</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/114">doi: 10.3390/automation7040114</a></p>
	<p>Authors:
		Emmanuel García-Galvan
		Antonio J. Cruz-Estrada
		Eduardo Vincent-Islas
		José R. Rivera-Ruiz
		Edson E. Cruz-Miguel
		Javier Calderón-Sánchez
		José R. García-Martínez
		</p>
	<p>High-precision X&amp;amp;ndash;Y positioning systems driven by stepper motors are widely used in industrial automation, manufacturing, and scientific instrumentation. However, fixed feedforward&amp;amp;ndash;feedback controllers may degrade when operating conditions vary, particularly as step frequency changes and the risk of synchronism loss increases. This work proposes an adaptive fuzzy feedforward&amp;amp;ndash;feedback controller for stepper-motor-driven X&amp;amp;ndash;Y positioning systems. The controller uses a Takagi&amp;amp;ndash;Sugeno (T&amp;amp;ndash;S) fuzzy inference system to adjust the proportional, derivative, and feedforward actions according to the tracking error, step frequency, and an auxiliary error-based adaptation variable. The control law is integrated with the inverse kinematics of the platform to generate synchronized step-domain commands, and a practical synchronism-preservation condition is established. Experimental validation on a NEMA 17-based X&amp;amp;ndash;Y platform showed accurate trajectory tracking, with a steady-state error of approximately 1.6[&amp;amp;mu;m] for a trapezoidal profile. For a multi-segment trajectory, the RMSE was 0.0749[mm] without load and 0.0760[mm] under a 7.5[kg] external load. Compared with a conventional PID controller, the proposed method reduced the RMSE from 0.1741[mm] to 0.0749[mm], while preserving motor synchronism.</p>
	]]></content:encoded>

	<dc:title>Adaptive Fuzzy Feedforward Compensation for High-Precision X&amp;amp;ndash;Y Positioning Systems Driven by Stepper Motors</dc:title>
			<dc:creator>Emmanuel García-Galvan</dc:creator>
			<dc:creator>Antonio J. Cruz-Estrada</dc:creator>
			<dc:creator>Eduardo Vincent-Islas</dc:creator>
			<dc:creator>José R. Rivera-Ruiz</dc:creator>
			<dc:creator>Edson E. Cruz-Miguel</dc:creator>
			<dc:creator>Javier Calderón-Sánchez</dc:creator>
			<dc:creator>José R. García-Martínez</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040114</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-07-23</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-07-23</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>114</prism:startingPage>
		<prism:doi>10.3390/automation7040114</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/114</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/113">

	<title>Automation, Vol. 7, Pages 113: Artificial Intelligence and Computer Vision for Intelligent Traffic Light Systems: A Systematic Review</title>
	<link>https://www.mdpi.com/2673-4052/7/4/113</link>
	<description>Urban traffic congestion remains a major barrier to sustainable mobility, necessitating a transition from static signaling to Intelligent Traffic Light Systems (ITLS). This study presents a systematic review in computer vision, artificial intelligence, and adaptive control algorithms for urban intersection optimization. Following a methodological framework informed by PRISMA guidelines, the study examines state-of-the-art architectures, including deep learning-based perception models and reinforcement learning agents. The findings indicate that, although two-stage detectors provide benchmark accuracy for vehicle perception, single-stage models and Vision Transformers offer the high-speed processing required for real-time traffic management. In addition, deep reinforcement learning enables autonomous, lane-specific optimization that outperforms traditional actuated and fixed-time systems. The review also identifies a persistent research gap in the deployment of these computational frameworks in the resource-constrained and heterogeneous infrastructures of developing countries. For the urban context of Riobamba, Ecuador, a phased implementation strategy is proposed that balances computational demands with the city&amp;amp;rsquo;s morphological and social characteristics. By bridging the gap between high-fidelity simulation and practical field deployment, this review provides a scalable framework for improving throughput, reducing emissions, and enhancing safety. Overall, these advances offer a promising pathway toward more resilient transportation systems in rapidly evolving Andean urban centers.</description>
	<pubDate>2026-07-23</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 113: Artificial Intelligence and Computer Vision for Intelligent Traffic Light Systems: A Systematic Review</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/113">doi: 10.3390/automation7040113</a></p>
	<p>Authors:
		Eugenia Naranjo
		Juan Diego Erazo Rodríguez
		Iván Sinaluisa
		Nestor Ulloa
		</p>
	<p>Urban traffic congestion remains a major barrier to sustainable mobility, necessitating a transition from static signaling to Intelligent Traffic Light Systems (ITLS). This study presents a systematic review in computer vision, artificial intelligence, and adaptive control algorithms for urban intersection optimization. Following a methodological framework informed by PRISMA guidelines, the study examines state-of-the-art architectures, including deep learning-based perception models and reinforcement learning agents. The findings indicate that, although two-stage detectors provide benchmark accuracy for vehicle perception, single-stage models and Vision Transformers offer the high-speed processing required for real-time traffic management. In addition, deep reinforcement learning enables autonomous, lane-specific optimization that outperforms traditional actuated and fixed-time systems. The review also identifies a persistent research gap in the deployment of these computational frameworks in the resource-constrained and heterogeneous infrastructures of developing countries. For the urban context of Riobamba, Ecuador, a phased implementation strategy is proposed that balances computational demands with the city&amp;amp;rsquo;s morphological and social characteristics. By bridging the gap between high-fidelity simulation and practical field deployment, this review provides a scalable framework for improving throughput, reducing emissions, and enhancing safety. Overall, these advances offer a promising pathway toward more resilient transportation systems in rapidly evolving Andean urban centers.</p>
	]]></content:encoded>

	<dc:title>Artificial Intelligence and Computer Vision for Intelligent Traffic Light Systems: A Systematic Review</dc:title>
			<dc:creator>Eugenia Naranjo</dc:creator>
			<dc:creator>Juan Diego Erazo Rodríguez</dc:creator>
			<dc:creator>Iván Sinaluisa</dc:creator>
			<dc:creator>Nestor Ulloa</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040113</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-07-23</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-07-23</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Systematic Review</prism:section>
	<prism:startingPage>113</prism:startingPage>
		<prism:doi>10.3390/automation7040113</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/113</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/112">

	<title>Automation, Vol. 7, Pages 112: Adaptive Sliding Mode Control for Robust Trajectory Tracking of Quadrotor UAVs Under Disturbances and Uncertainties</title>
	<link>https://www.mdpi.com/2673-4052/7/4/112</link>
	<description>This study presents an adaptive sliding mode control approach for trajectory tracking of a quadrotor unmanned aerial vehicle functioning under external interference and kinematic unpredictability. Linear and rotational motion equations formulated in inactive local coordinate system frames are developed with the aid of a nonlinear six-degrees-of-freedom quadrotor dynamic model. Through mitigating excessive switching activity, reliability is improved. Here, the proposed controller integrates sliding mode control with bounded adaptive switching gain factors and boundary-layer smoothing. The operational design is applied within a sequential outer-loop/inner-loop structure for linear and orientation control. The conventional sliding mode control, alongside the proportional derivative control, which employs MATLAB/Simulink R2024a simulations while being interference-affected with an unknown performance set-up, is deployed in this work to relatively appraise the proposed ASM controller. The assessment involves three-dimensional trajectory, control input characteristics, tracking error analysis, adaptive gain growth, and chattering analysis with quantitative performance metrics. The computational output revealed that the proposed ASMC attained superior tracking performance with limited oscillation and level control action. The controller achieves a total RMSE of approximately 0.38 m and a lower aggregate tracking error when using the conventional SMC and PD controllers under equivalent conditions. Furthermore, the adaptive gain mechanism successfully lowers chattering while maintaining robustness against interferences, a large amount of ambiguity, and inertial imbalance with signal noise. The results validate that the proposed ASMC delivers a functional balance between robustness, control smoothness, and tracking accuracy alongside execution homogeneity for autonomous quadrotor UAV trajectory tracking in unsettled and unstable environments.</description>
	<pubDate>2026-07-21</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 112: Adaptive Sliding Mode Control for Robust Trajectory Tracking of Quadrotor UAVs Under Disturbances and Uncertainties</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/112">doi: 10.3390/automation7040112</a></p>
	<p>Authors:
		Mukhtar Fatihu Hamza
		</p>
	<p>This study presents an adaptive sliding mode control approach for trajectory tracking of a quadrotor unmanned aerial vehicle functioning under external interference and kinematic unpredictability. Linear and rotational motion equations formulated in inactive local coordinate system frames are developed with the aid of a nonlinear six-degrees-of-freedom quadrotor dynamic model. Through mitigating excessive switching activity, reliability is improved. Here, the proposed controller integrates sliding mode control with bounded adaptive switching gain factors and boundary-layer smoothing. The operational design is applied within a sequential outer-loop/inner-loop structure for linear and orientation control. The conventional sliding mode control, alongside the proportional derivative control, which employs MATLAB/Simulink R2024a simulations while being interference-affected with an unknown performance set-up, is deployed in this work to relatively appraise the proposed ASM controller. The assessment involves three-dimensional trajectory, control input characteristics, tracking error analysis, adaptive gain growth, and chattering analysis with quantitative performance metrics. The computational output revealed that the proposed ASMC attained superior tracking performance with limited oscillation and level control action. The controller achieves a total RMSE of approximately 0.38 m and a lower aggregate tracking error when using the conventional SMC and PD controllers under equivalent conditions. Furthermore, the adaptive gain mechanism successfully lowers chattering while maintaining robustness against interferences, a large amount of ambiguity, and inertial imbalance with signal noise. The results validate that the proposed ASMC delivers a functional balance between robustness, control smoothness, and tracking accuracy alongside execution homogeneity for autonomous quadrotor UAV trajectory tracking in unsettled and unstable environments.</p>
	]]></content:encoded>

	<dc:title>Adaptive Sliding Mode Control for Robust Trajectory Tracking of Quadrotor UAVs Under Disturbances and Uncertainties</dc:title>
			<dc:creator>Mukhtar Fatihu Hamza</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040112</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-07-21</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-07-21</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>112</prism:startingPage>
		<prism:doi>10.3390/automation7040112</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/112</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/111">

	<title>Automation, Vol. 7, Pages 111: Intelligent Control of an Aeration Tank Using Model Predictive Control and a Digital Twin</title>
	<link>https://www.mdpi.com/2673-4052/7/4/111</link>
	<description>In the context of water scarcity and tightening environmental requirements, improving the energy efficiency of biological wastewater treatment processes has become particularly important. The aeration tank is one of the most energy-intensive and dynamically complex units, strongly affected by the variability in influent flow and composition. Conventional PID controllers do not provide predictive disturbance compensation and often result in excessive aeration and increased energy consumption. The study proposes an intelligent control approach based on a digital twin, neural network-based influent flow forecasting, and model predictive control (MPC). The digital twin represents a dynamic model of the biological process incorporating key state variables, including substrate, activated sludge, and dissolved oxygen concentrations. The LSTM neural network model is employed to predict the hydraulic load based on historical plant operation data, as well as to compensate for residual nonlinear dynamics that are not represented by the linearized MPC model. The predicted influent flow values are incorporated into the MPC framework as measured disturbances, enabling the generation of anticipatory control actions for the aeration system. The adequacy of the digital twin was validated using operational data from the wastewater treatment facilities of Semey city and was characterized by RMSE = 0.14 mg/L, MAE = 0.09 mg/L, R2 = 0.94, and MAPE = 6.3%. The simulation results demonstrated that, compared with the fuzzy PID controller, the application of MPC reduced the RMSE by 57.1%, decreased the overshoot from 24% to 8%, reduced the integral absolute error (IAE) by 60.9%, and lowered the energy consumption of the aeration system by 21.1%, while maintaining the dissolved oxygen concentration within the permissible operating range. The proposed Advisory MPC architecture is compatible with existing PLC&amp;amp;ndash;SCADA systems and can serve as a basis for the gradual digital modernization of wastewater treatment facilities without modifying the existing automation loops.</description>
	<pubDate>2026-07-20</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 111: Intelligent Control of an Aeration Tank Using Model Predictive Control and a Digital Twin</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/111">doi: 10.3390/automation7040111</a></p>
	<p>Authors:
		Alexandr Zolotov
		Tursynkhan Zhylkybayev
		Dinara Kozhakhmetova
		Yerbol Ospanov
		Bakhytgul Kopabayeva
		Rashid Nazarov
		Dmitriy Myassoyedov
		Tatyana Ustinova
		Kulken Zenkovich
		Ahmet Sakir Dokuz
		</p>
	<p>In the context of water scarcity and tightening environmental requirements, improving the energy efficiency of biological wastewater treatment processes has become particularly important. The aeration tank is one of the most energy-intensive and dynamically complex units, strongly affected by the variability in influent flow and composition. Conventional PID controllers do not provide predictive disturbance compensation and often result in excessive aeration and increased energy consumption. The study proposes an intelligent control approach based on a digital twin, neural network-based influent flow forecasting, and model predictive control (MPC). The digital twin represents a dynamic model of the biological process incorporating key state variables, including substrate, activated sludge, and dissolved oxygen concentrations. The LSTM neural network model is employed to predict the hydraulic load based on historical plant operation data, as well as to compensate for residual nonlinear dynamics that are not represented by the linearized MPC model. The predicted influent flow values are incorporated into the MPC framework as measured disturbances, enabling the generation of anticipatory control actions for the aeration system. The adequacy of the digital twin was validated using operational data from the wastewater treatment facilities of Semey city and was characterized by RMSE = 0.14 mg/L, MAE = 0.09 mg/L, R2 = 0.94, and MAPE = 6.3%. The simulation results demonstrated that, compared with the fuzzy PID controller, the application of MPC reduced the RMSE by 57.1%, decreased the overshoot from 24% to 8%, reduced the integral absolute error (IAE) by 60.9%, and lowered the energy consumption of the aeration system by 21.1%, while maintaining the dissolved oxygen concentration within the permissible operating range. The proposed Advisory MPC architecture is compatible with existing PLC&amp;amp;ndash;SCADA systems and can serve as a basis for the gradual digital modernization of wastewater treatment facilities without modifying the existing automation loops.</p>
	]]></content:encoded>

	<dc:title>Intelligent Control of an Aeration Tank Using Model Predictive Control and a Digital Twin</dc:title>
			<dc:creator>Alexandr Zolotov</dc:creator>
			<dc:creator>Tursynkhan Zhylkybayev</dc:creator>
			<dc:creator>Dinara Kozhakhmetova</dc:creator>
			<dc:creator>Yerbol Ospanov</dc:creator>
			<dc:creator>Bakhytgul Kopabayeva</dc:creator>
			<dc:creator>Rashid Nazarov</dc:creator>
			<dc:creator>Dmitriy Myassoyedov</dc:creator>
			<dc:creator>Tatyana Ustinova</dc:creator>
			<dc:creator>Kulken Zenkovich</dc:creator>
			<dc:creator>Ahmet Sakir Dokuz</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040111</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-07-20</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-07-20</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>111</prism:startingPage>
		<prism:doi>10.3390/automation7040111</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/111</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/110">

	<title>Automation, Vol. 7, Pages 110: Analysis and Comparison of Chebyshev&amp;ndash;Halley Multipoint Methods for Power Flow Calculation in Monopolar Direct-Current Networks</title>
	<link>https://www.mdpi.com/2673-4052/7/4/110</link>
	<description>The increasing penetration of direct-current (DC) technologies in power transmission and distribution systems necessitates efficient and robust tools for steady-state analysis. This paper presents a comparative evaluation of the Chebyshev&amp;amp;ndash;Halley (CH) family of multipoint iterative methods against the classical Newton&amp;amp;ndash;Raphson (NR) method for power flow calculation in monopolar DC networks. Both methods were implemented in MATLAB and tested on four radial test systems of increasing complexity (10, 21, 33, and 69 nodes) under three distinct initialization scenarios: optimal (flat start), adverse (V(0)=0.5 p.u.), and random (V(0)&amp;amp;sim;U[0.8,1.2] p.u.). Performance was assessed using key metrics including iteration count, CPU time, solution accuracy, and convergence failure rate. The results demonstrate that the cubic convergence of CH consistently reduces the number of iterations by one when compared to NR across all systems. However, this reduction does not translate into computational savings, as CH exhibits median CPU times 1.36 to 2.44 times higher than those of NR, given its higher cost per iteration, which involves solving two additional linear systems. Under adverse starting conditions, both methods converge for the 10-, 21-, and 33-node systems, but CH fails on the 69-node network due to severe Jacobian ill-conditioning, from which NR recovers through an implicit regularization mechanism. Under random initializations, both methods show high failure rates, reaching 100% in the 69-node network. It is concluded that, while CH offers superior convergence order and final accuracy, NR remains more computationally efficient for small- to medium-scale networks under flat-start conditions. The CH family is best justified in high-precision applications or larger networks where the iteration reduction may offset its per-step overhead. Future work should focus on extending CH to meshed and multi-source DC networks, developing quasi-Newton variants to reduce its computational cost, and designing hybrid NR-CH strategies that combine global robustness with local cubic convergence.</description>
	<pubDate>2026-07-19</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 110: Analysis and Comparison of Chebyshev&amp;ndash;Halley Multipoint Methods for Power Flow Calculation in Monopolar Direct-Current Networks</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/110">doi: 10.3390/automation7040110</a></p>
	<p>Authors:
		Sebastián Salazar-Méndez
		José Daniel Pico-Díaz
		Oscar Danilo Montoya
		</p>
	<p>The increasing penetration of direct-current (DC) technologies in power transmission and distribution systems necessitates efficient and robust tools for steady-state analysis. This paper presents a comparative evaluation of the Chebyshev&amp;amp;ndash;Halley (CH) family of multipoint iterative methods against the classical Newton&amp;amp;ndash;Raphson (NR) method for power flow calculation in monopolar DC networks. Both methods were implemented in MATLAB and tested on four radial test systems of increasing complexity (10, 21, 33, and 69 nodes) under three distinct initialization scenarios: optimal (flat start), adverse (V(0)=0.5 p.u.), and random (V(0)&amp;amp;sim;U[0.8,1.2] p.u.). Performance was assessed using key metrics including iteration count, CPU time, solution accuracy, and convergence failure rate. The results demonstrate that the cubic convergence of CH consistently reduces the number of iterations by one when compared to NR across all systems. However, this reduction does not translate into computational savings, as CH exhibits median CPU times 1.36 to 2.44 times higher than those of NR, given its higher cost per iteration, which involves solving two additional linear systems. Under adverse starting conditions, both methods converge for the 10-, 21-, and 33-node systems, but CH fails on the 69-node network due to severe Jacobian ill-conditioning, from which NR recovers through an implicit regularization mechanism. Under random initializations, both methods show high failure rates, reaching 100% in the 69-node network. It is concluded that, while CH offers superior convergence order and final accuracy, NR remains more computationally efficient for small- to medium-scale networks under flat-start conditions. The CH family is best justified in high-precision applications or larger networks where the iteration reduction may offset its per-step overhead. Future work should focus on extending CH to meshed and multi-source DC networks, developing quasi-Newton variants to reduce its computational cost, and designing hybrid NR-CH strategies that combine global robustness with local cubic convergence.</p>
	]]></content:encoded>

	<dc:title>Analysis and Comparison of Chebyshev&amp;amp;ndash;Halley Multipoint Methods for Power Flow Calculation in Monopolar Direct-Current Networks</dc:title>
			<dc:creator>Sebastián Salazar-Méndez</dc:creator>
			<dc:creator>José Daniel Pico-Díaz</dc:creator>
			<dc:creator>Oscar Danilo Montoya</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040110</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-07-19</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-07-19</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>110</prism:startingPage>
		<prism:doi>10.3390/automation7040110</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/110</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/109">

	<title>Automation, Vol. 7, Pages 109: Product-Assembly Planning Ontology for Integrating Product Design and Assembly Process Planning (APP)</title>
	<link>https://www.mdpi.com/2673-4052/7/4/109</link>
	<description>This paper presents a semantic approach to support knowledge sharing in the assembly domain. Specifically, it focuses on capturing and sharing assembly design knowledge and on integrating the assembly design domain with the Assembly Process Planning (APP) domain through ontological modeling. A multilayered, heavyweight ontology framework, called the Product-Assembly Planning Ontology (PAPO), is proposed to integrate product assembly and APP. The ontology is based on product assembly features and uses these design features to provide the high-level semantic knowledge necessary to integrate product assembly design with APP. The paper also describes a detailed methodology for ontology design. Additionally, a rule-based engine is developed to reason about the available assembly design and APP knowledge and to infer new knowledge from them. Case study examples are included to illustrate the approach.</description>
	<pubDate>2026-07-16</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 109: Product-Assembly Planning Ontology for Integrating Product Design and Assembly Process Planning (APP)</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/109">doi: 10.3390/automation7040109</a></p>
	<p>Authors:
		Baha M. Hasan
		Jan Wikander
		Mauro Onori
		</p>
	<p>This paper presents a semantic approach to support knowledge sharing in the assembly domain. Specifically, it focuses on capturing and sharing assembly design knowledge and on integrating the assembly design domain with the Assembly Process Planning (APP) domain through ontological modeling. A multilayered, heavyweight ontology framework, called the Product-Assembly Planning Ontology (PAPO), is proposed to integrate product assembly and APP. The ontology is based on product assembly features and uses these design features to provide the high-level semantic knowledge necessary to integrate product assembly design with APP. The paper also describes a detailed methodology for ontology design. Additionally, a rule-based engine is developed to reason about the available assembly design and APP knowledge and to infer new knowledge from them. Case study examples are included to illustrate the approach.</p>
	]]></content:encoded>

	<dc:title>Product-Assembly Planning Ontology for Integrating Product Design and Assembly Process Planning (APP)</dc:title>
			<dc:creator>Baha M. Hasan</dc:creator>
			<dc:creator>Jan Wikander</dc:creator>
			<dc:creator>Mauro Onori</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040109</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-07-16</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-07-16</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>109</prism:startingPage>
		<prism:doi>10.3390/automation7040109</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/109</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/108">

	<title>Automation, Vol. 7, Pages 108: A Hybrid Cognitive Radio and Multi-Agent Reinforcement Learning Framework for Jamming Resilience in Integrated FANET&amp;ndash;IoT&amp;ndash;IoV Systems</title>
	<link>https://www.mdpi.com/2673-4052/7/4/108</link>
	<description>Flying Ad-Hoc Networks (FANETs), Internet of Things (IoT), and Internet of Vehicles (IoV) are critical enablers of intelligent transportation and smart city ecosystems. Their reliance on shared wireless channels, however, exposes them to diverse jamming attacks that threaten communication reliability, mission effectiveness, and safety. This paper presents a comprehensive study of jamming threats in integrated FANET&amp;amp;ndash;IoT&amp;amp;ndash;IoV environments and analyzes conventional and advanced anti-jamming techniques across physical, link/MAC, spectral, spatial, temporal, and hybrid domains. To address the challenges posed by heterogeneous and dynamic network conditions, we propose a cross-layer anti-jamming framework that integrates Cognitive Radio (CR) for dynamic spectrum access and Multi-Agent Reinforcement Learning (MARL) for cooperative, adaptive decision-making. The framework employs a Perception Engine for local anomaly detection, a Cognitive Engine for constructing a collaborative jamming map, and a Decision and Action Engine for multi-agent DRL-based mitigation. Simulation results demonstrate that the proposed CR-MARL framework significantly improves packet delivery ratio, reduces latency, and adapts efficiently to varying jamming strategies, while maintaining low energy and computational overhead, making it suitable for resource-constrained UAVs, vehicles, and IoT sensors.</description>
	<pubDate>2026-07-10</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 108: A Hybrid Cognitive Radio and Multi-Agent Reinforcement Learning Framework for Jamming Resilience in Integrated FANET&amp;ndash;IoT&amp;ndash;IoV Systems</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/108">doi: 10.3390/automation7040108</a></p>
	<p>Authors:
		Rizwan Raza
		 Zahoor-ur-Rehman
		Muddasar Naeem
		Farhan Aadil
		Faheem Shehzad
		Antonio Coronato
		</p>
	<p>Flying Ad-Hoc Networks (FANETs), Internet of Things (IoT), and Internet of Vehicles (IoV) are critical enablers of intelligent transportation and smart city ecosystems. Their reliance on shared wireless channels, however, exposes them to diverse jamming attacks that threaten communication reliability, mission effectiveness, and safety. This paper presents a comprehensive study of jamming threats in integrated FANET&amp;amp;ndash;IoT&amp;amp;ndash;IoV environments and analyzes conventional and advanced anti-jamming techniques across physical, link/MAC, spectral, spatial, temporal, and hybrid domains. To address the challenges posed by heterogeneous and dynamic network conditions, we propose a cross-layer anti-jamming framework that integrates Cognitive Radio (CR) for dynamic spectrum access and Multi-Agent Reinforcement Learning (MARL) for cooperative, adaptive decision-making. The framework employs a Perception Engine for local anomaly detection, a Cognitive Engine for constructing a collaborative jamming map, and a Decision and Action Engine for multi-agent DRL-based mitigation. Simulation results demonstrate that the proposed CR-MARL framework significantly improves packet delivery ratio, reduces latency, and adapts efficiently to varying jamming strategies, while maintaining low energy and computational overhead, making it suitable for resource-constrained UAVs, vehicles, and IoT sensors.</p>
	]]></content:encoded>

	<dc:title>A Hybrid Cognitive Radio and Multi-Agent Reinforcement Learning Framework for Jamming Resilience in Integrated FANET&amp;amp;ndash;IoT&amp;amp;ndash;IoV Systems</dc:title>
			<dc:creator>Rizwan Raza</dc:creator>
			<dc:creator> Zahoor-ur-Rehman</dc:creator>
			<dc:creator>Muddasar Naeem</dc:creator>
			<dc:creator>Farhan Aadil</dc:creator>
			<dc:creator>Faheem Shehzad</dc:creator>
			<dc:creator>Antonio Coronato</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040108</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-07-10</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-07-10</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>108</prism:startingPage>
		<prism:doi>10.3390/automation7040108</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/108</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/107">

	<title>Automation, Vol. 7, Pages 107: UAV-Assisted MOSI/SOMI MIMO-FSO Relay for Resilient Transport Communication Links</title>
	<link>https://www.mdpi.com/2673-4052/7/4/107</link>
	<description>Reliable communication infrastructure is a fundamental component of Intelligent Transport Systems (ITSs), particularly in scenarios involving maritime corridors and emergency traffic management. In locations where optical fiber deployment is geographically constrained, unmanned aerial vehicle (UAV)-assisted free-space optical (FSO) relay links provide a flexible and rapidly deployable alternative. However, atmospheric attenuation, turbulence-induced fading, and wind-induced UAV misalignment can severely degrade link reliability and disrupt real-time transport data streams. This study proposes a payload-efficient multiple-input multiple-output free-space optical (MIMO-FSO) relay architecture based on a multi-output/single-input (MOSI) uplink and a single-output/multi-input (SOMI) downlink. Here, MOSI denotes multiple ground-based transmit apertures directed toward a single UAV receiving aperture, whereas SOMI denotes one UAV transmitting aperture serving multiple ground-based receiving apertures. Unlike conventional symmetric UAV-assisted MIMO-FSO relays that may duplicate diversity hardware on the aerial node, the proposed design shifts the parallel optical branches to the ground stations and keeps only one optical receiver and one optical transmitter on board the UAV. Under the adopted 4 × 4 comparison assumption, this reduces the UAV-side optical branch count from eight to two, corresponding to a 75% branch-count reduction proxy. System performance is evaluated over a 1.54 km relay link. The analytical framework describes Beer–Lambert attenuation, log-normal/gamma–gamma turbulence, and statistical pointing errors; in the OptiSystem implementation, their combined effects are represented by equivalent aggregate losses of 25 dB/km for atmospheric absorption/scattering and 25.5 dB/km for turbulence- and pointing-related degradation. Comparative simulations for SISO, 2 × 2, and 4 × 4 configurations show that the proposed 4 × 4 architecture increases the Q-factor from 8.38 to 18.25 and changes the OptiSystem-reported minimum BER from 2.73 × 10−17 to 9.95 × 10−75. Because a finite simulation cannot statistically validate error probabilities of this magnitude through raw error counting, values far below 10−12 are interpreted primarily as comparative indicators of receiver decision margin. The findings provide simulation-based evidence that the proposed architecture is a scalable candidate for resilient optical wireless backhaul in smart transport corridors under adverse propagation conditions.</description>
	<pubDate>2026-07-10</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 107: UAV-Assisted MOSI/SOMI MIMO-FSO Relay for Resilient Transport Communication Links</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/107">doi: 10.3390/automation7040107</a></p>
	<p>Authors:
		Ho Cuu
		Leminh Huynh
		Žarko Koboević
		</p>
	<p>Reliable communication infrastructure is a fundamental component of Intelligent Transport Systems (ITSs), particularly in scenarios involving maritime corridors and emergency traffic management. In locations where optical fiber deployment is geographically constrained, unmanned aerial vehicle (UAV)-assisted free-space optical (FSO) relay links provide a flexible and rapidly deployable alternative. However, atmospheric attenuation, turbulence-induced fading, and wind-induced UAV misalignment can severely degrade link reliability and disrupt real-time transport data streams. This study proposes a payload-efficient multiple-input multiple-output free-space optical (MIMO-FSO) relay architecture based on a multi-output/single-input (MOSI) uplink and a single-output/multi-input (SOMI) downlink. Here, MOSI denotes multiple ground-based transmit apertures directed toward a single UAV receiving aperture, whereas SOMI denotes one UAV transmitting aperture serving multiple ground-based receiving apertures. Unlike conventional symmetric UAV-assisted MIMO-FSO relays that may duplicate diversity hardware on the aerial node, the proposed design shifts the parallel optical branches to the ground stations and keeps only one optical receiver and one optical transmitter on board the UAV. Under the adopted 4 × 4 comparison assumption, this reduces the UAV-side optical branch count from eight to two, corresponding to a 75% branch-count reduction proxy. System performance is evaluated over a 1.54 km relay link. The analytical framework describes Beer–Lambert attenuation, log-normal/gamma–gamma turbulence, and statistical pointing errors; in the OptiSystem implementation, their combined effects are represented by equivalent aggregate losses of 25 dB/km for atmospheric absorption/scattering and 25.5 dB/km for turbulence- and pointing-related degradation. Comparative simulations for SISO, 2 × 2, and 4 × 4 configurations show that the proposed 4 × 4 architecture increases the Q-factor from 8.38 to 18.25 and changes the OptiSystem-reported minimum BER from 2.73 × 10−17 to 9.95 × 10−75. Because a finite simulation cannot statistically validate error probabilities of this magnitude through raw error counting, values far below 10−12 are interpreted primarily as comparative indicators of receiver decision margin. The findings provide simulation-based evidence that the proposed architecture is a scalable candidate for resilient optical wireless backhaul in smart transport corridors under adverse propagation conditions.</p>
	]]></content:encoded>

	<dc:title>UAV-Assisted MOSI/SOMI MIMO-FSO Relay for Resilient Transport Communication Links</dc:title>
			<dc:creator>Ho Cuu</dc:creator>
			<dc:creator>Leminh Huynh</dc:creator>
			<dc:creator>Žarko Koboević</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040107</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-07-10</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-07-10</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>107</prism:startingPage>
		<prism:doi>10.3390/automation7040107</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/107</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/106">

	<title>Automation, Vol. 7, Pages 106: Hybrid User Memory Filtering Algorithm for LLM-Based Knowledge Management Systems: Reducing Contextual Noise in Industrial Automation</title>
	<link>https://www.mdpi.com/2673-4052/7/4/106</link>
	<description>This paper presents a single-company field study on hybrid user memory filtering for Large Language Model (LLM)-based knowledge management systems, aiming to reduce contextual noise from irrelevant or outdated persistent memories. We propose the Hybrid Adaptive Filtering Engine (HAFE), which combines intent classification, ontology-based filtering, behavioral reuse prediction, and collaborative role-level comparison. HAFE was integrated into an industrial KM platform deployed at a major steel producer. In a field experiment with 120 engineers, HAFE reduced irrelevant memory retention by 41%, improved Mean Reciprocal Rank (MRR) by 12.5%, and increased user satisfaction (SUS score) by 18% (all p &amp;amp;lt; 0.01). The results suggest that proactive memory quality control can improve effectiveness and user experience in this specific industrial KMS setting, while further cross-domain validation is required.</description>
	<pubDate>2026-07-09</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 106: Hybrid User Memory Filtering Algorithm for LLM-Based Knowledge Management Systems: Reducing Contextual Noise in Industrial Automation</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/106">doi: 10.3390/automation7040106</a></p>
	<p>Authors:
		Viktor A. Vedeneev
		Viktor V. Kondratiev
		Konstantin V. Suslov
		Roman V. Kononenko
		Galina Yu. Vitkina
		Vitaliy A. Gladkikh
		Yulia I. Karlina
		Antonina I. Karlina
		</p>
	<p>This paper presents a single-company field study on hybrid user memory filtering for Large Language Model (LLM)-based knowledge management systems, aiming to reduce contextual noise from irrelevant or outdated persistent memories. We propose the Hybrid Adaptive Filtering Engine (HAFE), which combines intent classification, ontology-based filtering, behavioral reuse prediction, and collaborative role-level comparison. HAFE was integrated into an industrial KM platform deployed at a major steel producer. In a field experiment with 120 engineers, HAFE reduced irrelevant memory retention by 41%, improved Mean Reciprocal Rank (MRR) by 12.5%, and increased user satisfaction (SUS score) by 18% (all p &amp;amp;lt; 0.01). The results suggest that proactive memory quality control can improve effectiveness and user experience in this specific industrial KMS setting, while further cross-domain validation is required.</p>
	]]></content:encoded>

	<dc:title>Hybrid User Memory Filtering Algorithm for LLM-Based Knowledge Management Systems: Reducing Contextual Noise in Industrial Automation</dc:title>
			<dc:creator>Viktor A. Vedeneev</dc:creator>
			<dc:creator>Viktor V. Kondratiev</dc:creator>
			<dc:creator>Konstantin V. Suslov</dc:creator>
			<dc:creator>Roman V. Kononenko</dc:creator>
			<dc:creator>Galina Yu. Vitkina</dc:creator>
			<dc:creator>Vitaliy A. Gladkikh</dc:creator>
			<dc:creator>Yulia I. Karlina</dc:creator>
			<dc:creator>Antonina I. Karlina</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040106</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-07-09</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-07-09</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>106</prism:startingPage>
		<prism:doi>10.3390/automation7040106</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/106</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/105">

	<title>Automation, Vol. 7, Pages 105: Beyond Automation Levels: A Framework for Human&amp;ndash;Autonomy and Manned&amp;ndash;Unmanned Teaming</title>
	<link>https://www.mdpi.com/2673-4052/7/4/105</link>
	<description>Manned&amp;amp;ndash;unmanned teaming (MUMT) represents a critical evolution in collaborative operations across domains including search and rescue, firefighting, surveillance, and defense. Despite widespread interest in MUMT capabilities, the field lacks a unified taxonomy for classifying and comparing system capabilities, hindering systematic development and technology integration. This paper presents a comprehensive framework for MUMT that addresses the fundamental challenge of organizing and assessing cognitive agent capabilities within human&amp;amp;ndash;machine teams. Building upon established automation frameworks, we propose a three-dimensional framework comprising information analysis and inference, decision-making, and action execution. Each dimension defines six hierarchical levels of teaming, ranging from human-only operations to fully autonomous cognitive agent capabilities. The framework distinguishes itself from existing taxonomies by explicitly modeling collaborative teaming rather than simple task delegation, incorporating transparency requirements, and addressing dynamic authority relationships between humans and cognitive agents. The proposed taxonomy provides researchers and engineers with a common vocabulary for MUMT development, enables gap analysis for technology roadmaps, and facilitates the identification of integration opportunities across organizational boundaries.</description>
	<pubDate>2026-07-06</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 105: Beyond Automation Levels: A Framework for Human&amp;ndash;Autonomy and Manned&amp;ndash;Unmanned Teaming</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/105">doi: 10.3390/automation7040105</a></p>
	<p>Authors:
		Melina Athanasiadou
		Giovanni Franzini
		Adrien Metge
		</p>
	<p>Manned&amp;amp;ndash;unmanned teaming (MUMT) represents a critical evolution in collaborative operations across domains including search and rescue, firefighting, surveillance, and defense. Despite widespread interest in MUMT capabilities, the field lacks a unified taxonomy for classifying and comparing system capabilities, hindering systematic development and technology integration. This paper presents a comprehensive framework for MUMT that addresses the fundamental challenge of organizing and assessing cognitive agent capabilities within human&amp;amp;ndash;machine teams. Building upon established automation frameworks, we propose a three-dimensional framework comprising information analysis and inference, decision-making, and action execution. Each dimension defines six hierarchical levels of teaming, ranging from human-only operations to fully autonomous cognitive agent capabilities. The framework distinguishes itself from existing taxonomies by explicitly modeling collaborative teaming rather than simple task delegation, incorporating transparency requirements, and addressing dynamic authority relationships between humans and cognitive agents. The proposed taxonomy provides researchers and engineers with a common vocabulary for MUMT development, enables gap analysis for technology roadmaps, and facilitates the identification of integration opportunities across organizational boundaries.</p>
	]]></content:encoded>

	<dc:title>Beyond Automation Levels: A Framework for Human&amp;amp;ndash;Autonomy and Manned&amp;amp;ndash;Unmanned Teaming</dc:title>
			<dc:creator>Melina Athanasiadou</dc:creator>
			<dc:creator>Giovanni Franzini</dc:creator>
			<dc:creator>Adrien Metge</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040105</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-07-06</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-07-06</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>105</prism:startingPage>
		<prism:doi>10.3390/automation7040105</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/105</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/104">

	<title>Automation, Vol. 7, Pages 104: Application of Large Language Models for Detecting Semantic Ambiguity in Industrial Instructions: Impact on Human&amp;ndash;Machine Interaction and User Experience in Process Automation Systems of a Metallurgical Plant</title>
	<link>https://www.mdpi.com/2673-4052/7/4/104</link>
	<description>In the context of industrial digitalization and the widespread adoption of process automation systems, Knowledge Management Systems (KMS) play a key role in providing operational personnel with up-to-date instructions and regulations. However, the inherent ambiguity of natural language in technical documentation remains a serious obstacle, leading to incorrect operator actions, process deviations, and increased safety risks. This article investigates the integration of Large Language Models (LLMs) into KMS and its impact on user experience and human&amp;amp;ndash;machine interaction in industrial automation environments. A method called Semantic Latent Choice Detection is presented, designed to systematically identify interpretation ambiguities in process instructions and operator commands. Unlike existing approaches that require access to the internal model architecture (&amp;amp;ldquo;white box&amp;amp;rdquo;) or token-level logits, the proposed method is logit-free and operates with closed commercial LLMs (&amp;amp;ldquo;black box&amp;amp;rdquo;) via standard API interfaces. The method analyzes the semantic similarity of binary text blocks and polysemous terms within the context of a specific technological process. Using a metallurgical production case study, we demonstrate how the system detects hidden semantic collisions (e.g., the difference between &amp;amp;ldquo;adding ferroalloys into the ladle&amp;amp;rdquo; and &amp;amp;ldquo;feeding ferroalloys onto the conveyor&amp;amp;rdquo;) that are missed by traditional rule-based validation methods. Instead of arbitrarily selecting an interpretation, the system initiates a clarification request to the human operator, thereby reducing cognitive load, preventing erroneous automated decisions, and increasing trust in the KMS. An empirical evaluation conducted in a real-world industrial setting (unit control rooms and dispatch centers) shows a statistically significant reduction in errors related to misinterpretation of process regulations. The article contributes to the fields of automation engineering, knowledge management, and human-centered automation by proposing a novel method for validating operational instructions in high-risk industrial environments.</description>
	<pubDate>2026-07-05</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 104: Application of Large Language Models for Detecting Semantic Ambiguity in Industrial Instructions: Impact on Human&amp;ndash;Machine Interaction and User Experience in Process Automation Systems of a Metallurgical Plant</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/104">doi: 10.3390/automation7040104</a></p>
	<p>Authors:
		Viktor A. Vedeneev
		Viktor V. Kondratiev
		Konstantin V. Suslov
		Roman V. Kononenko
		Aleksey S. Govorkov
		Vitaliy A. Gladkikh
		Yulia I. Karlina
		Antonina I. Karlina
		</p>
	<p>In the context of industrial digitalization and the widespread adoption of process automation systems, Knowledge Management Systems (KMS) play a key role in providing operational personnel with up-to-date instructions and regulations. However, the inherent ambiguity of natural language in technical documentation remains a serious obstacle, leading to incorrect operator actions, process deviations, and increased safety risks. This article investigates the integration of Large Language Models (LLMs) into KMS and its impact on user experience and human&amp;amp;ndash;machine interaction in industrial automation environments. A method called Semantic Latent Choice Detection is presented, designed to systematically identify interpretation ambiguities in process instructions and operator commands. Unlike existing approaches that require access to the internal model architecture (&amp;amp;ldquo;white box&amp;amp;rdquo;) or token-level logits, the proposed method is logit-free and operates with closed commercial LLMs (&amp;amp;ldquo;black box&amp;amp;rdquo;) via standard API interfaces. The method analyzes the semantic similarity of binary text blocks and polysemous terms within the context of a specific technological process. Using a metallurgical production case study, we demonstrate how the system detects hidden semantic collisions (e.g., the difference between &amp;amp;ldquo;adding ferroalloys into the ladle&amp;amp;rdquo; and &amp;amp;ldquo;feeding ferroalloys onto the conveyor&amp;amp;rdquo;) that are missed by traditional rule-based validation methods. Instead of arbitrarily selecting an interpretation, the system initiates a clarification request to the human operator, thereby reducing cognitive load, preventing erroneous automated decisions, and increasing trust in the KMS. An empirical evaluation conducted in a real-world industrial setting (unit control rooms and dispatch centers) shows a statistically significant reduction in errors related to misinterpretation of process regulations. The article contributes to the fields of automation engineering, knowledge management, and human-centered automation by proposing a novel method for validating operational instructions in high-risk industrial environments.</p>
	]]></content:encoded>

	<dc:title>Application of Large Language Models for Detecting Semantic Ambiguity in Industrial Instructions: Impact on Human&amp;amp;ndash;Machine Interaction and User Experience in Process Automation Systems of a Metallurgical Plant</dc:title>
			<dc:creator>Viktor A. Vedeneev</dc:creator>
			<dc:creator>Viktor V. Kondratiev</dc:creator>
			<dc:creator>Konstantin V. Suslov</dc:creator>
			<dc:creator>Roman V. Kononenko</dc:creator>
			<dc:creator>Aleksey S. Govorkov</dc:creator>
			<dc:creator>Vitaliy A. Gladkikh</dc:creator>
			<dc:creator>Yulia I. Karlina</dc:creator>
			<dc:creator>Antonina I. Karlina</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040104</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-07-05</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-07-05</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>104</prism:startingPage>
		<prism:doi>10.3390/automation7040104</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/104</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/103">

	<title>Automation, Vol. 7, Pages 103: Lean Manufacturing Adaptation in High-Variety and Unstable Demand Engineer-to-Order Production: An Action Research Study Using Value Stream Mapping</title>
	<link>https://www.mdpi.com/2673-4052/7/4/103</link>
	<description>Engineer-to-Order (ETO) manufacturing environments are characterized by high product variety, low repetitiveness, and unstable demand, which pose significant challenges to the application of Lean Manufacturing (LM). This study investigates the application and adaptation of LM principles and tools in an ETO production line using an action research approach integrated with Value Stream Mapping (VSM). The research was conducted at a manufacturer of highly customized electrical equipment. An adapted method for calculating representative cycle times based on weighted production volumes was developed to support line sizing and workload balancing. The proposed future-state design incorporates multifunctional operators, FIFO lanes, daily scheduling, and pitch-based control. The results show a 9.5% reduction in labor requirements, a 61.7% decrease in manufacturing lead time, and a 75.0% reduction in overtime hours. Statistical validation using daily PPC records confirmed significant improvements in actual output, schedule adherence, overtime, and lead time after implementation. In addition to operational improvements, this study offers methodological contributions by proposing practical adaptations of LM tools suitable for high-variability ETO environments, thereby contributing to both theory and industrial practice.</description>
	<pubDate>2026-07-03</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 103: Lean Manufacturing Adaptation in High-Variety and Unstable Demand Engineer-to-Order Production: An Action Research Study Using Value Stream Mapping</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/103">doi: 10.3390/automation7040103</a></p>
	<p>Authors:
		Israel Galhardo
		José Antonio de Queiroz
		José Henrique de Freitas Gomes
		</p>
	<p>Engineer-to-Order (ETO) manufacturing environments are characterized by high product variety, low repetitiveness, and unstable demand, which pose significant challenges to the application of Lean Manufacturing (LM). This study investigates the application and adaptation of LM principles and tools in an ETO production line using an action research approach integrated with Value Stream Mapping (VSM). The research was conducted at a manufacturer of highly customized electrical equipment. An adapted method for calculating representative cycle times based on weighted production volumes was developed to support line sizing and workload balancing. The proposed future-state design incorporates multifunctional operators, FIFO lanes, daily scheduling, and pitch-based control. The results show a 9.5% reduction in labor requirements, a 61.7% decrease in manufacturing lead time, and a 75.0% reduction in overtime hours. Statistical validation using daily PPC records confirmed significant improvements in actual output, schedule adherence, overtime, and lead time after implementation. In addition to operational improvements, this study offers methodological contributions by proposing practical adaptations of LM tools suitable for high-variability ETO environments, thereby contributing to both theory and industrial practice.</p>
	]]></content:encoded>

	<dc:title>Lean Manufacturing Adaptation in High-Variety and Unstable Demand Engineer-to-Order Production: An Action Research Study Using Value Stream Mapping</dc:title>
			<dc:creator>Israel Galhardo</dc:creator>
			<dc:creator>José Antonio de Queiroz</dc:creator>
			<dc:creator>José Henrique de Freitas Gomes</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040103</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-07-03</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-07-03</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>103</prism:startingPage>
		<prism:doi>10.3390/automation7040103</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/103</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/102">

	<title>Automation, Vol. 7, Pages 102: FPGA-Compatible XSG Simulation of a Super-Twisting Sliding Mode Speed Control for a Dual-Star Induction Machine Using RFOC and MRAS Observer</title>
	<link>https://www.mdpi.com/2673-4052/7/4/102</link>
	<description>The control of Dual-Star Induction Machines (DSIMs) with high performance remains a challenging task, particularly in the presence of parameter variations and under sensorless operation. In practice, widely used controllers such as Proportional&amp;amp;ndash;Integral (PI) and classical sliding mode (SM) often reach their limits, especially in terms of dynamic responses, sensitivity to disturbances, and chattering, which can negatively affect system stability and efficiency. In this work, an improved Rotor Flux-Oriented Control (RFOC) strategy is proposed. It combines a super-twisting sliding mode (STSM) speed controller with a Model Reference Adaptive System (MRAS) observer. The STSM controller ensures faster convergence and enhanced robustness while significantly reducing chattering. Meanwhile, the MRAS observer enables accurate rotor speed estimation without mechanical sensors, thereby simplifying the system and improving reliability. The control scheme is developed using the Xilinx System Generator (XSG) in a fixed-point environment, providing an FPGA-oriented and compatible simulation framework. To assess its effectiveness, the proposed method is evaluated through several simulation scenarios and compared with conventional RFOC-PI and RFOC-SM approaches. The results demonstrate clear improvements in dynamic performance, disturbance rejection capability, and steady-state accuracy. Overall, the proposed approach provides a practical and efficient solution for DSIM drive systems operating under demanding conditions.</description>
	<pubDate>2026-07-01</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 102: FPGA-Compatible XSG Simulation of a Super-Twisting Sliding Mode Speed Control for a Dual-Star Induction Machine Using RFOC and MRAS Observer</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/102">doi: 10.3390/automation7040102</a></p>
	<p>Authors:
		Fatma Zohra Latrech
		Asma Ben Rhouma
		Adel Khedher
		</p>
	<p>The control of Dual-Star Induction Machines (DSIMs) with high performance remains a challenging task, particularly in the presence of parameter variations and under sensorless operation. In practice, widely used controllers such as Proportional&amp;amp;ndash;Integral (PI) and classical sliding mode (SM) often reach their limits, especially in terms of dynamic responses, sensitivity to disturbances, and chattering, which can negatively affect system stability and efficiency. In this work, an improved Rotor Flux-Oriented Control (RFOC) strategy is proposed. It combines a super-twisting sliding mode (STSM) speed controller with a Model Reference Adaptive System (MRAS) observer. The STSM controller ensures faster convergence and enhanced robustness while significantly reducing chattering. Meanwhile, the MRAS observer enables accurate rotor speed estimation without mechanical sensors, thereby simplifying the system and improving reliability. The control scheme is developed using the Xilinx System Generator (XSG) in a fixed-point environment, providing an FPGA-oriented and compatible simulation framework. To assess its effectiveness, the proposed method is evaluated through several simulation scenarios and compared with conventional RFOC-PI and RFOC-SM approaches. The results demonstrate clear improvements in dynamic performance, disturbance rejection capability, and steady-state accuracy. Overall, the proposed approach provides a practical and efficient solution for DSIM drive systems operating under demanding conditions.</p>
	]]></content:encoded>

	<dc:title>FPGA-Compatible XSG Simulation of a Super-Twisting Sliding Mode Speed Control for a Dual-Star Induction Machine Using RFOC and MRAS Observer</dc:title>
			<dc:creator>Fatma Zohra Latrech</dc:creator>
			<dc:creator>Asma Ben Rhouma</dc:creator>
			<dc:creator>Adel Khedher</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040102</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-07-01</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-07-01</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>102</prism:startingPage>
		<prism:doi>10.3390/automation7040102</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/102</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/101">

	<title>Automation, Vol. 7, Pages 101: A Taxonomy-Driven Analysis of Learning-Based Approaches in SLAM</title>
	<link>https://www.mdpi.com/2673-4052/7/4/101</link>
	<description>Learning-based approaches have significantly advanced the capabilities of Simultaneous Localization and Mapping (SLAM) systems, particularly in challenging environments characterized by noise, dynamic objects, and perceptual ambiguity. However, the literature remains highly heterogeneous in terms of sensing modalities, datasets, evaluation protocols, and reporting practices, making systematic comparison difficult. This paper presents a taxonomy-driven review of learning-based SLAM approaches, with particular emphasis on LiDAR-based systems in mobile robotics, and introduces a functional taxonomy that categorizes methods according to the role of learning within the SLAM architecture: (i) learning-enhanced front-end SLAM (T1), (ii) learning-enhanced back-end SLAM (T2), and (iii) learning-centric SLAM systems (T3). Representative studies were analyzed with respect to performance characteristics, robustness, computational requirements, datasets, and deployment-related evidence. The analysis shows that T1 approaches primarily improve local pose estimation and robustness, T2 methods enhance global consistency through learning-based loop closure and relocalization, and T3 approaches explore unified representations, semantic reasoning, and learning-centric autonomy, albeit with greater computational demands and limited deployment evidence. The review further indicates that hybrid approaches combining geometric and learning-based components constitute a prominent trend in the literature, frequently reporting improvements in accuracy and adaptability while maintaining compatibility with established SLAM frameworks. Nevertheless, these observations should be interpreted cautiously, as stronger empirical evidence for hybrid systems may partially reflect their greater technological maturity and broader evaluation history. Finally, the review identifies persistent challenges, including limited cross-domain generalization, high computational requirements, limited deployment-oriented evaluation, and the lack of standardized benchmarking and reporting practices. These findings highlight the need for more reproducible evaluation methodologies, uncertainty-aware learning strategies, and computationally efficient architectures for robust real-world autonomous SLAM.</description>
	<pubDate>2026-07-01</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 101: A Taxonomy-Driven Analysis of Learning-Based Approaches in SLAM</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/101">doi: 10.3390/automation7040101</a></p>
	<p>Authors:
		Rafael Rojas-Galván
		Luis F. Olmedo-García
		José R. García-Martínez
		José Manuel Alvarez-Alvarado
		Ricardo Rojas-Galván
		Juvenal Rodríguez-Reséndiz
		</p>
	<p>Learning-based approaches have significantly advanced the capabilities of Simultaneous Localization and Mapping (SLAM) systems, particularly in challenging environments characterized by noise, dynamic objects, and perceptual ambiguity. However, the literature remains highly heterogeneous in terms of sensing modalities, datasets, evaluation protocols, and reporting practices, making systematic comparison difficult. This paper presents a taxonomy-driven review of learning-based SLAM approaches, with particular emphasis on LiDAR-based systems in mobile robotics, and introduces a functional taxonomy that categorizes methods according to the role of learning within the SLAM architecture: (i) learning-enhanced front-end SLAM (T1), (ii) learning-enhanced back-end SLAM (T2), and (iii) learning-centric SLAM systems (T3). Representative studies were analyzed with respect to performance characteristics, robustness, computational requirements, datasets, and deployment-related evidence. The analysis shows that T1 approaches primarily improve local pose estimation and robustness, T2 methods enhance global consistency through learning-based loop closure and relocalization, and T3 approaches explore unified representations, semantic reasoning, and learning-centric autonomy, albeit with greater computational demands and limited deployment evidence. The review further indicates that hybrid approaches combining geometric and learning-based components constitute a prominent trend in the literature, frequently reporting improvements in accuracy and adaptability while maintaining compatibility with established SLAM frameworks. Nevertheless, these observations should be interpreted cautiously, as stronger empirical evidence for hybrid systems may partially reflect their greater technological maturity and broader evaluation history. Finally, the review identifies persistent challenges, including limited cross-domain generalization, high computational requirements, limited deployment-oriented evaluation, and the lack of standardized benchmarking and reporting practices. These findings highlight the need for more reproducible evaluation methodologies, uncertainty-aware learning strategies, and computationally efficient architectures for robust real-world autonomous SLAM.</p>
	]]></content:encoded>

	<dc:title>A Taxonomy-Driven Analysis of Learning-Based Approaches in SLAM</dc:title>
			<dc:creator>Rafael Rojas-Galván</dc:creator>
			<dc:creator>Luis F. Olmedo-García</dc:creator>
			<dc:creator>José R. García-Martínez</dc:creator>
			<dc:creator>José Manuel Alvarez-Alvarado</dc:creator>
			<dc:creator>Ricardo Rojas-Galván</dc:creator>
			<dc:creator>Juvenal Rodríguez-Reséndiz</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040101</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-07-01</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-07-01</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Systematic Review</prism:section>
	<prism:startingPage>101</prism:startingPage>
		<prism:doi>10.3390/automation7040101</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/101</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/4/100">

	<title>Automation, Vol. 7, Pages 100: An A*-Distance-Guided Exploration Strategy for Multi-AGV Path Planning</title>
	<link>https://www.mdpi.com/2673-4052/7/4/100</link>
	<description>A common limitation of existing multi-AGV cooperative systems is their reliance on the obstacle-agnostic Manhattan distance as the basis for reward signals. This causes agents to receive misleading feedback, engage in excessive futile exploration, and ultimately achieve poor training quality. To address this, we introduce an A*-distance guidance mechanism for multi-agent reinforcement learning (MARL) path planning, built on the precise path distance computed via the A* algorithm (A*-distance). Within the QMIX framework, we incorporate an A*-distance-based guiding function into the action selection mechanism. This function evaluates candidate actions by quantifying their immediate effect on the A*-distance, providing positive incentives for actions that bring the agent closer to the goal and applying negative penalties for those that lead it farther away. This effectively biases exploration towards actions that genuinely shorten the obstacle-aware path to the goal, suppresses ineffective exploration, and accelerates policy convergence. Experiments in four warehouse environments (simple obstacles, complex obstacles, large-scale, and congested) show that, compared with standard QMIX, the proposed method achieves higher global average reward and faster convergence. The advantage grows as environment scale and obstacle density increase. In the large-scale and congested environments, standard QMIX and the other MARL baselines fail to solve the task, whereas the proposed method still succeeds. It is the only learning-based method to solve these hardest tasks while keeping path length close to that of dedicated search-based solvers. Ablation experiments further show that the A*-distance-guided action selection is the primary contributor to these gains, while the A*-distance reward plays a supporting role.</description>
	<pubDate>2026-06-25</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 100: An A*-Distance-Guided Exploration Strategy for Multi-AGV Path Planning</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/4/100">doi: 10.3390/automation7040100</a></p>
	<p>Authors:
		Ying Zhou
		Yixin Feng
		Peiyan Mao
		Pengfei Wang
		</p>
	<p>A common limitation of existing multi-AGV cooperative systems is their reliance on the obstacle-agnostic Manhattan distance as the basis for reward signals. This causes agents to receive misleading feedback, engage in excessive futile exploration, and ultimately achieve poor training quality. To address this, we introduce an A*-distance guidance mechanism for multi-agent reinforcement learning (MARL) path planning, built on the precise path distance computed via the A* algorithm (A*-distance). Within the QMIX framework, we incorporate an A*-distance-based guiding function into the action selection mechanism. This function evaluates candidate actions by quantifying their immediate effect on the A*-distance, providing positive incentives for actions that bring the agent closer to the goal and applying negative penalties for those that lead it farther away. This effectively biases exploration towards actions that genuinely shorten the obstacle-aware path to the goal, suppresses ineffective exploration, and accelerates policy convergence. Experiments in four warehouse environments (simple obstacles, complex obstacles, large-scale, and congested) show that, compared with standard QMIX, the proposed method achieves higher global average reward and faster convergence. The advantage grows as environment scale and obstacle density increase. In the large-scale and congested environments, standard QMIX and the other MARL baselines fail to solve the task, whereas the proposed method still succeeds. It is the only learning-based method to solve these hardest tasks while keeping path length close to that of dedicated search-based solvers. Ablation experiments further show that the A*-distance-guided action selection is the primary contributor to these gains, while the A*-distance reward plays a supporting role.</p>
	]]></content:encoded>

	<dc:title>An A*-Distance-Guided Exploration Strategy for Multi-AGV Path Planning</dc:title>
			<dc:creator>Ying Zhou</dc:creator>
			<dc:creator>Yixin Feng</dc:creator>
			<dc:creator>Peiyan Mao</dc:creator>
			<dc:creator>Pengfei Wang</dc:creator>
		<dc:identifier>doi: 10.3390/automation7040100</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-06-25</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-06-25</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>4</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>100</prism:startingPage>
		<prism:doi>10.3390/automation7040100</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/4/100</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/3/99">

	<title>Automation, Vol. 7, Pages 99: A Novel Genetic Algorithm for the Dual-Resource Flexible Job Shop Scheduling Problem with Partial Resource Allocation</title>
	<link>https://www.mdpi.com/2673-4052/7/3/99</link>
	<description>This paper proposes a genetic algorithm (GA) for the Dual-Resource Flexible Job Shop Scheduling Problem with Partial Resource Allocation (DRFJSSP-PRA), a particular variant of a dual-resource constrained scheduling problem that has not been fully explored due to its intricate nature. The DRFJSSP-PRA poses a challenging scheduling problem, having several applications in many industries, including food, chemistry and pharmaceutics. The proposed algorithm is applied to real-world scheduling instances in pharmaceutical quality control. The objective function considered is the total completion time. The GA is compared with three state-of-the-art algorithms. For small- and medium-size instances, the proposed algorithm achieves optimal or near optimal results for the majority of the instances tested. For large-sized instances, the proposed GA outperforms all the other algorithms, in all of the tested instances. Thus, the experimental results show that the proposed GA achieves competitive results for any type of instance. The proposed algorithm also has the ability to optimize production processes through scheduling, leading to potential cost savings, increased efficiency, and improved competitiveness.</description>
	<pubDate>2026-06-20</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 99: A Novel Genetic Algorithm for the Dual-Resource Flexible Job Shop Scheduling Problem with Partial Resource Allocation</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/3/99">doi: 10.3390/automation7030099</a></p>
	<p>Authors:
		Diogo Marta
		Bernardo Firme
		Miguel S. E. Martins
		João M. C. Sousa
		Susana M. Vieira
		</p>
	<p>This paper proposes a genetic algorithm (GA) for the Dual-Resource Flexible Job Shop Scheduling Problem with Partial Resource Allocation (DRFJSSP-PRA), a particular variant of a dual-resource constrained scheduling problem that has not been fully explored due to its intricate nature. The DRFJSSP-PRA poses a challenging scheduling problem, having several applications in many industries, including food, chemistry and pharmaceutics. The proposed algorithm is applied to real-world scheduling instances in pharmaceutical quality control. The objective function considered is the total completion time. The GA is compared with three state-of-the-art algorithms. For small- and medium-size instances, the proposed algorithm achieves optimal or near optimal results for the majority of the instances tested. For large-sized instances, the proposed GA outperforms all the other algorithms, in all of the tested instances. Thus, the experimental results show that the proposed GA achieves competitive results for any type of instance. The proposed algorithm also has the ability to optimize production processes through scheduling, leading to potential cost savings, increased efficiency, and improved competitiveness.</p>
	]]></content:encoded>

	<dc:title>A Novel Genetic Algorithm for the Dual-Resource Flexible Job Shop Scheduling Problem with Partial Resource Allocation</dc:title>
			<dc:creator>Diogo Marta</dc:creator>
			<dc:creator>Bernardo Firme</dc:creator>
			<dc:creator>Miguel S. E. Martins</dc:creator>
			<dc:creator>João M. C. Sousa</dc:creator>
			<dc:creator>Susana M. Vieira</dc:creator>
		<dc:identifier>doi: 10.3390/automation7030099</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-06-20</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-06-20</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>3</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>99</prism:startingPage>
		<prism:doi>10.3390/automation7030099</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/3/99</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/3/98">

	<title>Automation, Vol. 7, Pages 98: Hybrid NMPC-ESO-PINSE Approach for Liquid Level Control in a Nonlinear Four-Tank System: Integration of Deep Learning and Extended State Observation Under Stochastic Uncertainties</title>
	<link>https://www.mdpi.com/2673-4052/7/3/98</link>
	<description>Liquid storage tanks are widely used in sectors such as water treatment, oil and gas, food processing, and chemical manufacturing. Knowing the exact amount of liquid in a tank is essential for ensuring safety, preventing spills, and optimizing process control; therefore, the liquid level in a tank must be maintained at a precise reference point. This is where liquid level control for tanks becomes crucial and constitutes a fundamental problem in the industrial sector due to nonlinearities, multivariable coupling, and stochastic disturbances. Given the drawbacks of available control methods, such as classical Model Predictive Control (MPC), which are highly dependent on model accuracy and struggle to reject complex stochastic noise, predicting random disturbances represents a major technological challenge. A new approach is proposed to specifically address the problem and challenge of the four-tank system, where water levels in two lower tanks must be controlled by two pumps, often with varying delays and significant parameter disturbances. To establish a relationship between expected performance and MPC parameters, this approach uses a novel hybrid nonlinear MPC, Extended State Observer, and Physics-Informed Neural State Estimation (NMPC-ESO-PINSE) architecture. A Physics-Informed Neural State Estimation (PINSE) layer, chosen for its learning capacity, is designed to filter sensor noise by applying Bernoulli&amp;amp;rsquo;s physical laws, while an Extended State Observer (ESO) is integrated to capture and compensate for unmodeled uncertainties in the process. Finally, a proposed hybrid (NMPC-ESO-PINSE) strategy leverages these clean, physically consistent state estimations to solve a non-convex optimization problem via Sequential Quadratic Programming (SQP), computing optimal pump voltages. Extensive numerical simulations demonstrate the superior resilience of this decoupled framework against parametric drifts and continuous noise sequences, yielding a +27.36% reduction in global Root Mean Square Error (RMSE) compared to standard NMPC, accelerating the closed-loop settling time to 15.2&amp;amp;nbsp;s, and restricting transient overshoot to just 0.18%.</description>
	<pubDate>2026-06-18</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 98: Hybrid NMPC-ESO-PINSE Approach for Liquid Level Control in a Nonlinear Four-Tank System: Integration of Deep Learning and Extended State Observation Under Stochastic Uncertainties</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/3/98">doi: 10.3390/automation7030098</a></p>
	<p>Authors:
		Zohra Zidane
		El Mostafa Atify
		Mohammed Zidane
		Ahmed Boumezzough
		</p>
	<p>Liquid storage tanks are widely used in sectors such as water treatment, oil and gas, food processing, and chemical manufacturing. Knowing the exact amount of liquid in a tank is essential for ensuring safety, preventing spills, and optimizing process control; therefore, the liquid level in a tank must be maintained at a precise reference point. This is where liquid level control for tanks becomes crucial and constitutes a fundamental problem in the industrial sector due to nonlinearities, multivariable coupling, and stochastic disturbances. Given the drawbacks of available control methods, such as classical Model Predictive Control (MPC), which are highly dependent on model accuracy and struggle to reject complex stochastic noise, predicting random disturbances represents a major technological challenge. A new approach is proposed to specifically address the problem and challenge of the four-tank system, where water levels in two lower tanks must be controlled by two pumps, often with varying delays and significant parameter disturbances. To establish a relationship between expected performance and MPC parameters, this approach uses a novel hybrid nonlinear MPC, Extended State Observer, and Physics-Informed Neural State Estimation (NMPC-ESO-PINSE) architecture. A Physics-Informed Neural State Estimation (PINSE) layer, chosen for its learning capacity, is designed to filter sensor noise by applying Bernoulli&amp;amp;rsquo;s physical laws, while an Extended State Observer (ESO) is integrated to capture and compensate for unmodeled uncertainties in the process. Finally, a proposed hybrid (NMPC-ESO-PINSE) strategy leverages these clean, physically consistent state estimations to solve a non-convex optimization problem via Sequential Quadratic Programming (SQP), computing optimal pump voltages. Extensive numerical simulations demonstrate the superior resilience of this decoupled framework against parametric drifts and continuous noise sequences, yielding a +27.36% reduction in global Root Mean Square Error (RMSE) compared to standard NMPC, accelerating the closed-loop settling time to 15.2&amp;amp;nbsp;s, and restricting transient overshoot to just 0.18%.</p>
	]]></content:encoded>

	<dc:title>Hybrid NMPC-ESO-PINSE Approach for Liquid Level Control in a Nonlinear Four-Tank System: Integration of Deep Learning and Extended State Observation Under Stochastic Uncertainties</dc:title>
			<dc:creator>Zohra Zidane</dc:creator>
			<dc:creator>El Mostafa Atify</dc:creator>
			<dc:creator>Mohammed Zidane</dc:creator>
			<dc:creator>Ahmed Boumezzough</dc:creator>
		<dc:identifier>doi: 10.3390/automation7030098</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-06-18</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-06-18</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>3</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>98</prism:startingPage>
		<prism:doi>10.3390/automation7030098</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/3/98</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/3/97">

	<title>Automation, Vol. 7, Pages 97: Data-Driven State Estimation for Nonlinear Stochastic Systems Using Gaussian Process-Based Adaptive Interacting Multiple Model Particle Filtering</title>
	<link>https://www.mdpi.com/2673-4052/7/3/97</link>
	<description>This paper focuses on state estimation for nonlinear stochastic systems with multiple switching models, especially under challenging conditions where the model dynamics are unknown and the transition probability matrix is uniformly distributed. Gaussian process regression is employed to learn the unknown system dynamics from an offline discrete dataset and is integrated into an interacting multiple model particle filtering framework. GPR enables data-driven learning of both state transition and observation functions. To cope with model uncertainty and uninformative prior transition knowledge, particularly under uniformly initialized TPM, a dual-layer adaptive TPM update strategy based on hidden Markov model inference is further incorporated. Finally, the proposed method is validated through simulations and compared with IMMPF under different assumptions on system dynamics and TPMs. The results show that, even without prior knowledge of the system dynamics or precise TPM information, the proposed GP-AIMMPF maintains robust and accurate state estimation performance.</description>
	<pubDate>2026-06-18</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 97: Data-Driven State Estimation for Nonlinear Stochastic Systems Using Gaussian Process-Based Adaptive Interacting Multiple Model Particle Filtering</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/3/97">doi: 10.3390/automation7030097</a></p>
	<p>Authors:
		Xueqi Yuan
		Qing Sun
		</p>
	<p>This paper focuses on state estimation for nonlinear stochastic systems with multiple switching models, especially under challenging conditions where the model dynamics are unknown and the transition probability matrix is uniformly distributed. Gaussian process regression is employed to learn the unknown system dynamics from an offline discrete dataset and is integrated into an interacting multiple model particle filtering framework. GPR enables data-driven learning of both state transition and observation functions. To cope with model uncertainty and uninformative prior transition knowledge, particularly under uniformly initialized TPM, a dual-layer adaptive TPM update strategy based on hidden Markov model inference is further incorporated. Finally, the proposed method is validated through simulations and compared with IMMPF under different assumptions on system dynamics and TPMs. The results show that, even without prior knowledge of the system dynamics or precise TPM information, the proposed GP-AIMMPF maintains robust and accurate state estimation performance.</p>
	]]></content:encoded>

	<dc:title>Data-Driven State Estimation for Nonlinear Stochastic Systems Using Gaussian Process-Based Adaptive Interacting Multiple Model Particle Filtering</dc:title>
			<dc:creator>Xueqi Yuan</dc:creator>
			<dc:creator>Qing Sun</dc:creator>
		<dc:identifier>doi: 10.3390/automation7030097</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-06-18</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-06-18</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>3</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>97</prism:startingPage>
		<prism:doi>10.3390/automation7030097</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/3/97</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/3/96">

	<title>Automation, Vol. 7, Pages 96: Rights-Based AI in Cyber&amp;ndash;Physical Systems: A Governance Framework for Socio-Technical Resilience and Trust</title>
	<link>https://www.mdpi.com/2673-4052/7/3/96</link>
	<description>AI-enabled cyber&amp;amp;ndash;physical systems (CPSs) are increasingly deployed in public governance contexts where they sense human populations, infer classifications or risks, and trigger interventions that can shape liberty, equality, and access to essential services. In these deployments, governance failures often arise not only from model error but from systems-level interactions across data generation, model updates, organizational practices, and downstream actuation. This paper introduces a Risk&amp;amp;ndash;Rights&amp;amp;ndash;Rules (3R) architecture that treats fundamental rights and legal rules as enforceable constraints on the sensing&amp;amp;ndash;inference&amp;amp;ndash;actuation loop, rather than as external ethical aspirations. Building on established risk-management baselines and safety engineering practice, we specify a testable assurance object, a structured 3R assurance case, that links rights claims to explicit assumptions, measurable evidence, and accountable control points across the lifecycle. The approach is designed to reduce &amp;amp;ldquo;legitimacy drift&amp;amp;rdquo; in stochastic decision pipelines by making uncertainty, demographic error, contestability, and procurement leverage auditable at the system level. The result is a governance blueprint for high-consequence public-sector AI deployments for governance failures, which is both technically robust and institutionally defensible.</description>
	<pubDate>2026-06-15</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 96: Rights-Based AI in Cyber&amp;ndash;Physical Systems: A Governance Framework for Socio-Technical Resilience and Trust</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/3/96">doi: 10.3390/automation7030096</a></p>
	<p>Authors:
		Maral Niazi
		Hossein Hassani
		Madison Lee
		</p>
	<p>AI-enabled cyber&amp;amp;ndash;physical systems (CPSs) are increasingly deployed in public governance contexts where they sense human populations, infer classifications or risks, and trigger interventions that can shape liberty, equality, and access to essential services. In these deployments, governance failures often arise not only from model error but from systems-level interactions across data generation, model updates, organizational practices, and downstream actuation. This paper introduces a Risk&amp;amp;ndash;Rights&amp;amp;ndash;Rules (3R) architecture that treats fundamental rights and legal rules as enforceable constraints on the sensing&amp;amp;ndash;inference&amp;amp;ndash;actuation loop, rather than as external ethical aspirations. Building on established risk-management baselines and safety engineering practice, we specify a testable assurance object, a structured 3R assurance case, that links rights claims to explicit assumptions, measurable evidence, and accountable control points across the lifecycle. The approach is designed to reduce &amp;amp;ldquo;legitimacy drift&amp;amp;rdquo; in stochastic decision pipelines by making uncertainty, demographic error, contestability, and procurement leverage auditable at the system level. The result is a governance blueprint for high-consequence public-sector AI deployments for governance failures, which is both technically robust and institutionally defensible.</p>
	]]></content:encoded>

	<dc:title>Rights-Based AI in Cyber&amp;amp;ndash;Physical Systems: A Governance Framework for Socio-Technical Resilience and Trust</dc:title>
			<dc:creator>Maral Niazi</dc:creator>
			<dc:creator>Hossein Hassani</dc:creator>
			<dc:creator>Madison Lee</dc:creator>
		<dc:identifier>doi: 10.3390/automation7030096</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-06-15</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-06-15</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>3</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>96</prism:startingPage>
		<prism:doi>10.3390/automation7030096</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/3/96</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/3/95">

	<title>Automation, Vol. 7, Pages 95: Feedforward Neural Network-Based MPC Optimized by Hybrid Fractional PSO&amp;ndash;SQP for Trajectory Tracking of Autonomous Vehicles</title>
	<link>https://www.mdpi.com/2673-4052/7/3/95</link>
	<description>Background/Objective: Autonomous vehicles (AVs) require control algorithms capable of handling complex and dynamic environments while satisfying multiple conflicting objectives such as safety, comfort, energy efficiency, and trajectory accuracy. Model predictive control (MPC) offers a principled framework for multi-constraint optimization, yet its real-time feasibility remains challenging for nonlinear vehicle dynamics. Methods: This paper presents a feedforward neural network (FNN)-based MPC framework for autonomous vehicle trajectory tracking. The FNN approximates the coupled vehicle dynamics and visual preview error model using an algebraic sum of log-sigmoid functions. Three adaptive FNN parameter sets, namely, the scaling factor, convergence parameter, and time-shifting parameter, are jointly optimized using a hybrid algorithm that combines the global search capability of fractional particle swarm optimization (FPSO) with the local refinement of sequential quadratic programming (SQP). Results: Comprehensive scenario-based simulations are performed to evaluate trajectory tracking dynamics under dry conditions with an adhesion coefficient of 0.8 and a vehicle mass of 1723 kg moving at a speed of 80 km/h. The results are quantitatively compared with a traditional PID controller and a structurally comparable MPC framework from the literature under identical simulation conditions; related DRL- and RL-based methods are discussed qualitatively for contextual orientation only. The stability, reliability, and computational complexity of the proposed framework are examined based on the mean square error, fitness value, and computational budget in GFLOPs for 100 independent runs. Conclusions: The proposed FNN-based MPC framework demonstrates improved tracking accuracy and optimizer reliability in simulation. While the present results indicate promising computational behavior, real-time deployment will require further validation on embedded automotive hardware and under closed-loop real-time constraints.</description>
	<pubDate>2026-06-15</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 95: Feedforward Neural Network-Based MPC Optimized by Hybrid Fractional PSO&amp;ndash;SQP for Trajectory Tracking of Autonomous Vehicles</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/3/95">doi: 10.3390/automation7030095</a></p>
	<p>Authors:
		Fahad Alotaibi
		Habib Dhahri
		Saleh Almohaimeed
		Awais Mahmood
		</p>
	<p>Background/Objective: Autonomous vehicles (AVs) require control algorithms capable of handling complex and dynamic environments while satisfying multiple conflicting objectives such as safety, comfort, energy efficiency, and trajectory accuracy. Model predictive control (MPC) offers a principled framework for multi-constraint optimization, yet its real-time feasibility remains challenging for nonlinear vehicle dynamics. Methods: This paper presents a feedforward neural network (FNN)-based MPC framework for autonomous vehicle trajectory tracking. The FNN approximates the coupled vehicle dynamics and visual preview error model using an algebraic sum of log-sigmoid functions. Three adaptive FNN parameter sets, namely, the scaling factor, convergence parameter, and time-shifting parameter, are jointly optimized using a hybrid algorithm that combines the global search capability of fractional particle swarm optimization (FPSO) with the local refinement of sequential quadratic programming (SQP). Results: Comprehensive scenario-based simulations are performed to evaluate trajectory tracking dynamics under dry conditions with an adhesion coefficient of 0.8 and a vehicle mass of 1723 kg moving at a speed of 80 km/h. The results are quantitatively compared with a traditional PID controller and a structurally comparable MPC framework from the literature under identical simulation conditions; related DRL- and RL-based methods are discussed qualitatively for contextual orientation only. The stability, reliability, and computational complexity of the proposed framework are examined based on the mean square error, fitness value, and computational budget in GFLOPs for 100 independent runs. Conclusions: The proposed FNN-based MPC framework demonstrates improved tracking accuracy and optimizer reliability in simulation. While the present results indicate promising computational behavior, real-time deployment will require further validation on embedded automotive hardware and under closed-loop real-time constraints.</p>
	]]></content:encoded>

	<dc:title>Feedforward Neural Network-Based MPC Optimized by Hybrid Fractional PSO&amp;amp;ndash;SQP for Trajectory Tracking of Autonomous Vehicles</dc:title>
			<dc:creator>Fahad Alotaibi</dc:creator>
			<dc:creator>Habib Dhahri</dc:creator>
			<dc:creator>Saleh Almohaimeed</dc:creator>
			<dc:creator>Awais Mahmood</dc:creator>
		<dc:identifier>doi: 10.3390/automation7030095</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-06-15</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-06-15</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>3</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>95</prism:startingPage>
		<prism:doi>10.3390/automation7030095</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/3/95</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
        <item rdf:about="https://www.mdpi.com/2673-4052/7/3/94">

	<title>Automation, Vol. 7, Pages 94: SLAM-Based Autonomous CO2 Mapping for Indoor Environmental Monitoring: A Proof-of-Concept Framework for Multi-Parameter Hazard Assessment</title>
	<link>https://www.mdpi.com/2673-4052/7/3/94</link>
	<description>Environmental monitoring in hazardous indoor zones conventionally relies on fixed-sensor networks or manual inspections, both of which suffer from spatial blind spots and increased human exposure risks. This paper addresses the problem of transforming sparse, mobile sensor measurements into spatially resolved risk assessments in GPS-denied environments. We propose a Hazard Index (HI) framework that normalizes environmental parameters against established safety thresholds into a unified, graduated risk metric with O(N) computational complexity, where N is the number of monitored parameters. The framework is designed for multi-parameter hazard assessment; the present work validates the computational pipeline, spatial mapping methodology, and classification logic through single-parameter CO2 detection (N=1) deployed on a LiDAR-guided robotic platform integrating an MQ-135 gas sensor interfaced via a NodeMCU ESP8266 microcontroller. Experimental validation across a 144 sq ft indoor area achieved a trajectory-following RMSE of 0.54 ft relative to planned waypoints using Hector SLAM without odometry, detected CO2 concentrations ranging from 0.02% to 0.25%, and identified a hazardous region encompassing eight measurement points (HI&amp;amp;ge;1.0) using a three-tier classification scheme (Safe, Elevated, Hazardous) within 225 s of active mapping. The framework provides a lightweight computational footprint suitable for real-time evaluation on an NVIDIA Jetson Nano. The proposed approach establishes a cost-effective, reproducible methodology for autonomous indoor environmental monitoring, with the modular architecture designed for future expansion to multi-parameter sensing.</description>
	<pubDate>2026-06-15</pubDate>

	<content:encoded><![CDATA[
	<p><b>Automation, Vol. 7, Pages 94: SLAM-Based Autonomous CO2 Mapping for Indoor Environmental Monitoring: A Proof-of-Concept Framework for Multi-Parameter Hazard Assessment</b></p>
	<p>Automation <a href="https://www.mdpi.com/2673-4052/7/3/94">doi: 10.3390/automation7030094</a></p>
	<p>Authors:
		Prajakta Salunkhe
		Mahesh Shirole
		Ninad Mehendale
		</p>
	<p>Environmental monitoring in hazardous indoor zones conventionally relies on fixed-sensor networks or manual inspections, both of which suffer from spatial blind spots and increased human exposure risks. This paper addresses the problem of transforming sparse, mobile sensor measurements into spatially resolved risk assessments in GPS-denied environments. We propose a Hazard Index (HI) framework that normalizes environmental parameters against established safety thresholds into a unified, graduated risk metric with O(N) computational complexity, where N is the number of monitored parameters. The framework is designed for multi-parameter hazard assessment; the present work validates the computational pipeline, spatial mapping methodology, and classification logic through single-parameter CO2 detection (N=1) deployed on a LiDAR-guided robotic platform integrating an MQ-135 gas sensor interfaced via a NodeMCU ESP8266 microcontroller. Experimental validation across a 144 sq ft indoor area achieved a trajectory-following RMSE of 0.54 ft relative to planned waypoints using Hector SLAM without odometry, detected CO2 concentrations ranging from 0.02% to 0.25%, and identified a hazardous region encompassing eight measurement points (HI&amp;amp;ge;1.0) using a three-tier classification scheme (Safe, Elevated, Hazardous) within 225 s of active mapping. The framework provides a lightweight computational footprint suitable for real-time evaluation on an NVIDIA Jetson Nano. The proposed approach establishes a cost-effective, reproducible methodology for autonomous indoor environmental monitoring, with the modular architecture designed for future expansion to multi-parameter sensing.</p>
	]]></content:encoded>

	<dc:title>SLAM-Based Autonomous CO2 Mapping for Indoor Environmental Monitoring: A Proof-of-Concept Framework for Multi-Parameter Hazard Assessment</dc:title>
			<dc:creator>Prajakta Salunkhe</dc:creator>
			<dc:creator>Mahesh Shirole</dc:creator>
			<dc:creator>Ninad Mehendale</dc:creator>
		<dc:identifier>doi: 10.3390/automation7030094</dc:identifier>
	<dc:source>Automation</dc:source>
	<dc:date>2026-06-15</dc:date>

	<prism:publicationName>Automation</prism:publicationName>
	<prism:publicationDate>2026-06-15</prism:publicationDate>
	<prism:volume>7</prism:volume>
	<prism:number>3</prism:number>
	<prism:section>Article</prism:section>
	<prism:startingPage>94</prism:startingPage>
		<prism:doi>10.3390/automation7030094</prism:doi>
	<prism:url>https://www.mdpi.com/2673-4052/7/3/94</prism:url>

	<cc:license rdf:resource="CC BY 4.0"/>
</item>
    
<cc:License rdf:about="https://creativecommons.org/licenses/by/4.0/">
	<cc:permits rdf:resource="https://creativecommons.org/ns#Reproduction" />
	<cc:permits rdf:resource="https://creativecommons.org/ns#Distribution" />
	<cc:permits rdf:resource="https://creativecommons.org/ns#DerivativeWorks" />
</cc:License>

</rdf:RDF>
