Sensors

14 pages, 881 KiB

Open AccessArticle

Integrating Machine Learning for Predictive Maintenance on Resource-Constrained PLCs: A Feasibility Study

by Riccardo Mennilli, Luigi Mazza and Andrea Mura

Sensors 2025, 25(2), 537; https://doi.org/10.3390/s25020537 (registering DOI) - 17 Jan 2025

This study investigates the potential of deploying a neural network model on an advanced programmable logic controller (PLC), specifically the Finder Opta™, for real-time inference within the predictive maintenance framework. In the context of Industry 4.0, edge computing aims to process data directly [...] Read more.

This study investigates the potential of deploying a neural network model on an advanced programmable logic controller (PLC), specifically the Finder Opta™, for real-time inference within the predictive maintenance framework. In the context of Industry 4.0, edge computing aims to process data directly on local devices rather than relying on a cloud infrastructure. This approach minimizes latency, enhances data security, and reduces the bandwidth required for data transmission, making it ideal for industrial applications that demand immediate response times. Despite the limited memory and processing power inherent to many edge devices, this proof-of-concept demonstrates the suitability of the Finder Opta™ for such applications. Using acoustic data, a convolutional neural network (CNN) is deployed to infer the rotational speed of a mechanical test bench. The findings underscore the potential of the Finder Opta™ to support scalable and efficient predictive maintenance solutions, laying the groundwork for future research in real-time anomaly detection. By enabling machine learning capabilities on compact, resource-constrained hardware, this approach promises a cost-effective, adaptable solution for diverse industrial environments. Full article

(This article belongs to the Special Issue Advanced Sensors in Nondestructive Testing and Structural Health Monitoring)

17 pages, 4281 KiB

Open AccessArticle

GravelSens: a Smart Gravel Sensor for High-Resolution, Non-Destructive Monitoring of Clogging Dynamics

by Kaan Koca, Eckhard Schleicher, André Bieberle, Stefan Haun, Silke Wieprecht and Markus Noack

Sensors 2025, 25(2), 536; https://doi.org/10.3390/s25020536 (registering DOI) - 17 Jan 2025

Abstract

Engineers, geomorphologists, and ecologists acknowledge the need for temporally and spatially resolved measurements of sediment clogging (also known as colmation) in permeable gravel-bed rivers due to its adverse impacts on water and habitat quality. In this paper, we present a novel method for [...] Read more.

Engineers, geomorphologists, and ecologists acknowledge the need for temporally and spatially resolved measurements of sediment clogging (also known as colmation) in permeable gravel-bed rivers due to its adverse impacts on water and habitat quality. In this paper, we present a novel method for non-destructive, real-time measurements of pore-scale sediment deposition and monitoring of clogging by using wire-mesh sensors (WMSs) embedded in spheres, forming a smart gravel bed (GravelSens). The measuring principle is based on one-by-one voltage excitation of transmitter electrodes, followed by simultaneous measurements of the resulting current by receiver electrodes at each crossing measuring pores. The currents are then linked to the conductive component of fluid impedance. The measurement performance of the developed sensor is validated by applying the Maxwell Garnett and parallel models to sensor data and comparing the results to data obtained by gamma ray computed tomography (CT). GravelSens is tested and validated under varying filling conditions of different particle sizes ranging from sand to fine gravel. The close agreement between GravelSens and CT measurements indicates the technology’s applicability in sediment–water research while also suggesting its potential for other solid–liquid two-phase flows. This pore-scale measurement and visualization system offers the capability to monitor clogging and de-clogging dynamics within pore spaces up to 10,000 Hz, making it the first laboratory equipment capable of performing such in situ measurements without radiation. Thus, GravelSens is a major improvement over existing methods and holds promise for advancing the understanding of flow–sediment–ecology interactions. Full article

(This article belongs to the Section Environmental Sensing)

20 pages, 7198 KiB

Open AccessArticle

Virtual Node-Driven Cloud–Edge Collaborative Resource Scheduling for Surveillance with Visual Sensors

by Xinyang Gu, Zhansheng Duan, Guangyuan Ye and Zhenjun Chang

Sensors 2025, 25(2), 535; https://doi.org/10.3390/s25020535 (registering DOI) - 17 Jan 2025

Abstract

For public security purposes, distributed surveillance systems are widely deployed in key areas. These systems comprise visual sensors, edge computing boxes, and cloud servers. Resource scheduling algorithms are critical to ensure such systems’ robustness and efficiency. They balance workloads and need to meet [...] Read more.

For public security purposes, distributed surveillance systems are widely deployed in key areas. These systems comprise visual sensors, edge computing boxes, and cloud servers. Resource scheduling algorithms are critical to ensure such systems’ robustness and efficiency. They balance workloads and need to meet real-time monitoring and emergency response requirements. Existing works have primarily focused on optimizing Quality of Service (QoS), latency, and energy consumption in edge computing under resource constraints. However, the issue of task congestion due to insufficient physical resources has been rarely investigated. In this paper, we tackle the challenges posed by large workloads and limited resources in the context of surveillance with visual sensors. First, we introduce the concept of virtual nodes for managing resource shortages, referred to as virtual node-driven resource scheduling. Then, we propose a convex-objective integer linear programming (ILP) model based on this concept and demonstrate its efficiency. Additionally, we propose three alternative virtual node-driven scheduling algorithms, the extension of a random algorithm, a genetic algorithm, and a heuristic algorithm, respectively. These algorithms serve as benchmarks for comparison with the proposed ILP model. Experimental results show that all the scheduling algorithms can effectively address the challenge of offloading multiple priority tasks under resource constraints. Furthermore, the ILP model shows the best scheduling performance among them. Full article

(This article belongs to the Special Issue Cloud and Edge Computing for IoT Applications)

15 pages, 3496 KiB

Open AccessArticle

PC-CS-YOLO: High-Precision Obstacle Detection for Visually Impaired Safety

by Jincheng Li, Menglin Zheng, Danyang Dong and Xing Xie

Sensors 2025, 25(2), 534; https://doi.org/10.3390/s25020534 (registering DOI) - 17 Jan 2025

Abstract

The issue of obstacle avoidance and safety for visually impaired individuals has been a major topic of research. However, complex street environments still pose significant challenges for blind obstacle detection systems. Existing solutions often fail to provide real-time, accurate obstacle avoidance decisions. In [...] Read more.

The issue of obstacle avoidance and safety for visually impaired individuals has been a major topic of research. However, complex street environments still pose significant challenges for blind obstacle detection systems. Existing solutions often fail to provide real-time, accurate obstacle avoidance decisions. In this study, we propose a blind obstacle detection system based on the PC-CS-YOLO model. The system improves the backbone network by adopting the partial convolutional feed-forward network (PCFN) to reduce computational redundancy. Additionally, to enhance the network’s robustness in multi-scale feature fusion, we introduce the Cross-Scale Attention Fusion (CSAF) mechanism, which integrates features from different sensory domains to achieve superior performance. Compared to state-of-the-art networks, our system shows improvements of 2.0%, 3.9%, and 1.5% in precision, recall, and mAP50, respectively. When evaluated on a GPU, the inference speed is 20.6 ms, which is 15.3 ms faster than YOLO11, meeting the real-time requirements for blind obstacle avoidance systems. Full article

(This article belongs to the Special Issue Image Processing and Pattern Recognition Based on Deep Learning—2nd Edition)

13 pages, 566 KiB

Open AccessArticle

Monitoring Sleep and Nightly Recovery with Wrist-Worn Wearables: Links to Training Load and Performance Adaptations

by Olli-Pekka Nuuttila, Daniela Schäfer Olstad, Kaisu Martinmäki, Arja Uusitalo and Heikki Kyröläinen

Sensors 2025, 25(2), 533; https://doi.org/10.3390/s25020533 (registering DOI) - 17 Jan 2025

Abstract

Previous studies on the effects of intensified training on sleep quality/quantity have been somewhat contradictory. Moreover, recreational athletes often track various sleep metrics, and those metrics’ actual connections to training adaptations are unknown. This study explored the effects of intensified training on sleep [...] Read more.

Previous studies on the effects of intensified training on sleep quality/quantity have been somewhat contradictory. Moreover, recreational athletes often track various sleep metrics, and those metrics’ actual connections to training adaptations are unknown. This study explored the effects of intensified training on sleep and nightly recovery along with their associations with training adaptations. A total of 24 participants (10 females) performed a 3-week baseline training period (BL), a 2-week overload period (OL), and a 1-week recovery period (REC), which were followed by test days (T1–T3). The endurance performance was assessed with a 3000 m running test. Throughout all of the periods, the nightly recovery information was monitored with a wrist-worn wearable, including sleep quantity and quality, heart rate (HR) and HR variability (HRV), and proprietary parameters combining several parameters and scaling the results individually. In addition, the perceived strain and muscle soreness were evaluated daily. The 3000 m running performance improved from T1 to T2 (−1.2 ± 1.7%, p = 0.006) and from T1 to T3 (−1.7 ± 1.2%, p = 0.002). The perceived strain and muscle soreness increased (p < 0.001) from the final week of the BL to the final week of the OL, but the subjective sleep quality and nightly recovery metrics remained unchanged. The OL average of the proprietary parameter, autonomic nervous system charge (“ANS charge”, combining the HR, HRV, and breathing rate), as well as the change in the sleep HR and HRV from the BL to the OL, were associated (p < 0.05) with a change in the 3000 m running time. In conclusion, the subjective recovery metrics were impaired by intensified training, while the sleep and nightly recovery metrics showed no consistent changes. However, there were substantial interindividual differences in nightly recovery, which were also associated with the training adaptations. Therefore, monitoring nightly recovery can help in recognizing individual responses to training and assist in optimizing training prescriptions. Full article

(This article belongs to the Special Issue Sensors in Sports)

16 pages, 2965 KiB

Open AccessArticle

Symmetry Breaking as a Basis for Characterization of Dielectric Materials

by Dubravko Tomić and Zvonimir Šipuš

Sensors 2025, 25(2), 532; https://doi.org/10.3390/s25020532 (registering DOI) - 17 Jan 2025

Abstract

This paper introduces a novel method for measuring the dielectric permittivity of materials within the microwave and millimeter wave frequency ranges. The proposed approach, classified as a guided wave transmission system, employs a periodic transmission line structure characterized by mirror/glide symmetry. The dielectric [...] Read more.

This paper introduces a novel method for measuring the dielectric permittivity of materials within the microwave and millimeter wave frequency ranges. The proposed approach, classified as a guided wave transmission system, employs a periodic transmission line structure characterized by mirror/glide symmetry. The dielectric permittivity is deduced by measuring the transmission properties of such structure when presence of the dielectric material breaks the inherent symmetry of the structure and consequently introduce a stopband in propagation characteristic. To explore the influence of symmetry breaking on propagation properties, an analytical dispersion equation, for both symmetries, is formulated using the Rigorous Coupled Wave Analysis (RCWA) combined with the matrix transverse resonance condition. Based on the analytical equation, an optimization procedure and linearized model for a sensing structure is obtained, specifically for X-band characterization of FR4 substrates. The theoretical results of the model are validated with full wave simulations and experimentally. Full article

(This article belongs to the Special Issue Feature Papers in Physical Sensors 2024)

► Show Figures

Figure 1

46 pages, 615 KiB

Open AccessReview

A Comprehensive Survey of Deep Learning Approaches in Image Processing

by Maria Trigka and Elias Dritsas

Sensors 2025, 25(2), 531; https://doi.org/10.3390/s25020531 (registering DOI) - 17 Jan 2025

Abstract

The integration of deep learning (DL) into image processing has driven transformative advancements, enabling capabilities far beyond the reach of traditional methodologies. This survey offers an in-depth exploration of the DL approaches that have redefined image processing, tracing their evolution from early innovations [...] Read more.

The integration of deep learning (DL) into image processing has driven transformative advancements, enabling capabilities far beyond the reach of traditional methodologies. This survey offers an in-depth exploration of the DL approaches that have redefined image processing, tracing their evolution from early innovations to the latest state-of-the-art developments. It also analyzes the progression of architectural designs and learning paradigms that have significantly enhanced the ability to process and interpret complex visual data. Key advancements, such as techniques improving model efficiency, generalization, and robustness, are examined, showcasing DL’s ability to address increasingly sophisticated image-processing tasks across diverse domains. Metrics used for rigorous model evaluation are also discussed, underscoring the importance of performance assessment in varied application contexts. The impact of DL in image processing is highlighted through its ability to tackle complex challenges and generate actionable insights. Finally, this survey identifies potential future directions, including the integration of emerging technologies like quantum computing and neuromorphic architectures for enhanced efficiency and federated learning for privacy-preserving training. Additionally, it highlights the potential of combining DL with emerging technologies such as edge computing and explainable artificial intelligence (AI) to address scalability and interpretability challenges. These advancements are positioned to further extend the capabilities and applications of DL, driving innovation in image processing. Full article

(This article belongs to the Special Issue Deep Learning-Based Image and Signal Sensing and Processing: 2nd Edition)

► Show Figures

Figure 1

24 pages, 7093 KiB

Open AccessArticle

Comparison of Manual, Automatic, and Voice Control in Wheelchair Navigation Simulation in Virtual Environments: Performance Evaluation of User and Motion Sickness

by Enrique Antonio Pedroza-Santiago, José Emilio Quiroz-Ibarra, Erik René Bojorges-Valdez and Miguel Ángel Padilla-Castañeda

Sensors 2025, 25(2), 530; https://doi.org/10.3390/s25020530 (registering DOI) - 17 Jan 2025

Abstract

Mobility is essential for individuals with physical disabilities, and wheelchairs significantly enhance their quality of life. Recent advancements focus on developing sophisticated control systems for effective and efficient interaction. This study evaluates the usability and performance of three wheelchair control modes manual, automatic, [...] Read more.

Mobility is essential for individuals with physical disabilities, and wheelchairs significantly enhance their quality of life. Recent advancements focus on developing sophisticated control systems for effective and efficient interaction. This study evaluates the usability and performance of three wheelchair control modes manual, automatic, and voice controlled using a virtual reality (VR) simulation tool. VR provides a controlled and repeatable environment to assess navigation performance and motion sickness across three scenarios: supermarket, museum, and city. Twenty participants completed nine tests each, resulting in 180 trials. Findings revealed significant differences in navigation efficiency, distance, and collision rates across control modes and scenarios. Automatic control consistently achieved faster navigation times and fewer collisions, particularly in the supermarket. Manual control offered precision but required greater user effort. Voice control, while intuitive, resulted in longer distances traveled and higher collision rates in complex scenarios like the city. Motion sickness levels varied across scenarios, with higher discomfort reported in the city during voice and automatic control. Participant feedback, gathered via a Likert scale questionnaire, highlighted the potential of VR simulation for evaluating user comfort and performance. This research underscores the advantages of VR-based testing for rapid prototyping and user-centered design, offering valuable insights into improving wheelchair control systems. Future work will explore adaptive algorithms to enhance usability and accessibility in real world applications. Full article

(This article belongs to the Topic Simulations and Applications of Augmented and Virtual Reality, 2nd Edition)

► Show Figures

Figure 1

19 pages, 137082 KiB

Open AccessArticle

Classification and Monitoring of Salt Marsh Vegetation in the Yellow River Delta Based on Multi-Source Remote Sensing Data Fusion

by Ran Xu, Yanguo Fan, Bowen Fan, Guangyue Feng and Ruotong Li

Sensors 2025, 25(2), 529; https://doi.org/10.3390/s25020529 (registering DOI) - 17 Jan 2025

Abstract

Salt marsh vegetation in the Yellow River Delta, including Phragmites australis (P. australis), Suaeda salsa (S. salsa), and Tamarix chinensis (T. chinensis), is essential for the stability of wetland ecosystems. In recent years, salt marsh vegetation has [...] Read more.

Salt marsh vegetation in the Yellow River Delta, including Phragmites australis (P. australis), Suaeda salsa (S. salsa), and Tamarix chinensis (T. chinensis), is essential for the stability of wetland ecosystems. In recent years, salt marsh vegetation has experienced severe degradation, which is primarily due to invasive species and human activities. Therefore, the accurate monitoring of the spatial distribution of these vegetation types is critical for the ecological protection and restoration of the Yellow River Delta. This study proposes a multi-source remote sensing data fusion method based on Sentinel-1 and Sentinel-2 imagery, integrating the temporal characteristics of optical and SAR (synthetic aperture radar) data for the classification mapping of salt marsh vegetation in the Yellow River Delta. Phenological and polarization features were extracted to capture vegetation characteristics. A random forest algorithm was then applied to evaluate the impact of different feature combinations on classification accuracy. Combining optical and SAR time-series data significantly enhanced classification accuracy, particularly in differentiating P. australis, S. salsa, and T. chinensis. The integration of phenological features, polarization ratio, and polarization difference achieved a classification accuracy of 93.51% with a Kappa coefficient of 0.917, outperforming the use of individual data sources. Full article

(This article belongs to the Section Remote Sensors)

► Show Figures

Figure 1

17 pages, 1807 KiB

Open AccessArticle

Spectrum Sensing in Very Low SNR Environment Using Multi-Scale Temporal Correlation Perception with Residual Attention

by Song Hong and Weiqiang Xu

Sensors 2025, 25(2), 528; https://doi.org/10.3390/s25020528 (registering DOI) - 17 Jan 2025

Abstract

Spectrum sensing is recognized as a viable strategy to alleviate the scarcity of spectrum resources and to optimize their usage. In this paper, considering the time-varying characteristics and the dependence on various timescales within a time series of samples composed of in-phase (I) [...] Read more.

Spectrum sensing is recognized as a viable strategy to alleviate the scarcity of spectrum resources and to optimize their usage. In this paper, considering the time-varying characteristics and the dependence on various timescales within a time series of samples composed of in-phase (I) and quadrature (Q) component signals, we propose a multi-scale time-correlated perceptual attention model named MSTC-PANet. The model consists of multiple parallel temporal correlation perceptual attention (TCPA) modules, enabling us to extract features at different timescales and identify dependencies among features across various timescales. Our simulations show that MSTC-PANet significantly improves the detection of channel occupancy at low signal-to-noise ratios (SNR), particularly in untrained scenarios with lower SNR conditions and modulation uncertainties. The analysis of the ROC curve indicates that at an SNR of -20 dB, the proposed MSTC-PANet achieves a detection rate of 98% with a false alarm rate of 10%. Furthermore, MSTC-PANet, which has been trained using digital modulation techniques, also demonstrates applicability to analog modulation. Full article

(This article belongs to the Special Issue Intelligent Sensor Fusion and AI Applications in Wireless Communication Networks)

► Show Figures

Figure 1

18 pages, 5011 KiB

Open AccessArticle

Improving Industrial Quality Control: A Transfer Learning Approach to Surface Defect Detection

by Ângela Semitela, Miguel Pereira, António Completo, Nuno Lau and José P. Santos

Sensors 2025, 25(2), 527; https://doi.org/10.3390/s25020527 (registering DOI) - 17 Jan 2025

Abstract

To automate the quality control of painted surfaces of heating devices, an automatic defect detection and classification system was developed by combining deflectometry and bright light-based illumination on the image acquisition, deep learning models for the classification of non-defective (OK) and defective (NOK) [...] Read more.

To automate the quality control of painted surfaces of heating devices, an automatic defect detection and classification system was developed by combining deflectometry and bright light-based illumination on the image acquisition, deep learning models for the classification of non-defective (OK) and defective (NOK) surfaces that fused dual-modal information at the decision level, and an online network for information dispatching and visualization. Three decision-making algorithms were tested for implementation: a new model built and trained from scratch and transfer learning of pre-trained networks (ResNet-50 and Inception V3). The results revealed that the two illumination modes employed widened the type of defects that could be identified with this system, while maintaining its lower computational complexity by performing multi-modal fusion at the decision level. Furthermore, the pre-trained networks achieved higher accuracies on defect classification compared to the self-built network, with ResNet-50 displaying higher accuracy. The inspection system consistently obtained fast and accurate surface classifications because it imposed OK classification on models trained with images from both illumination modes. The obtained surface information was then successfully sent to a server to be forwarded to a graphical user interface for visualization. The developed system showed considerable robustness, demonstrating its potential as an efficient tool for industrial quality control. Full article

(This article belongs to the Section Industrial Sensors)

► Show Figures

Figure 1

21 pages, 12015 KiB

Open AccessArticle

Segment Any Leaf 3D: A Zero-Shot 3D Leaf Instance Segmentation Method Based on Multi-View Images

by Yunlong Wang and Zhiyong Zhang

Sensors 2025, 25(2), 526; https://doi.org/10.3390/s25020526 (registering DOI) - 17 Jan 2025

Abstract

Exploring the relationships between plant phenotypes and genetic information requires advanced phenotypic analysis techniques for precise characterization. However, the diversity and variability of plant morphology challenge existing methods, which often fail to generalize across species and require extensive annotated data, especially for 3D [...] Read more.

Exploring the relationships between plant phenotypes and genetic information requires advanced phenotypic analysis techniques for precise characterization. However, the diversity and variability of plant morphology challenge existing methods, which often fail to generalize across species and require extensive annotated data, especially for 3D datasets. This paper proposes a zero-shot 3D leaf instance segmentation method using RGB sensors. It extends the 2D segmentation model SAM (Segment Anything Model) to 3D through a multi-view strategy. RGB image sequences captured from multiple viewpoints are used to reconstruct 3D plant point clouds via multi-view stereo. HQ-SAM (High-Quality Segment Anything Model) segments leaves in 2D, and the segmentation is mapped to the 3D point cloud. An incremental fusion method based on confidence scores aggregates results from different views into a final output. Evaluated on a custom peanut seedling dataset, the method achieved point-level precision, recall, and F1 scores over 0.9 and object-level mIoU and precision above 0.75 under two IoU thresholds. The results show that the method achieves state-of-the-art segmentation quality while offering zero-shot capability and generalizability, demonstrating significant potential in plant phenotyping. Full article

(This article belongs to the Section Smart Agriculture)

► Show Figures

Figure 1

15 pages, 11786 KiB

Open AccessArticle

Improving the Selectivity of Metal Oxide Semiconductor Sensors for Mustard Gas Simulant 2-Chloroethyl Ethyl Sulfide by Combining the Laminated Structure and Temperature Dynamic Modulation

by Yadong Liu, Siyue Zhao, Lijuan You, Yong Xu, Renjun Si and Shunping Zhang

Sensors 2025, 25(2), 525; https://doi.org/10.3390/s25020525 (registering DOI) - 17 Jan 2025

Abstract

Insufficient selectivity is a major constraint to the further development of metal oxide semiconductor (MOS) sensors for chemical warfare agents, and this paper proposed an improved scheme combining catalytic layer/gas-sensitive layer laminated structure with temperature dynamic modulation for the Mustard gas (HD) MOS [...] Read more.

Insufficient selectivity is a major constraint to the further development of metal oxide semiconductor (MOS) sensors for chemical warfare agents, and this paper proposed an improved scheme combining catalytic layer/gas-sensitive layer laminated structure with temperature dynamic modulation for the Mustard gas (HD) MOS sensor. Mustard gas simulant 2-Chloroethyl ethyl sulfide (2-CEES) was used as the target gas, (Pt + Pd + Rh)@Al₂O₃ as the catalytic layer material, (Pt + Rh)@WO₃ as the gas-sensitive layer material, the (Pt + Pd + Rh)@Al₂O₃/(Pt + Rh)@WO₃ sensor was prepared, and the sensor was tested for 2-CEES and 12 battlefield environment simulation gases under temperature dynamic modulation. The results showed that the sensor only showed obvious characteristic peaks in the resistance response curves to HD under certain conditions (100–400 °C, the highest temperature was held for 1 s and the lowest temperature was held for 2 s), and its peak height reached 6.12, which was far higher than other gases, thus realizing the high selectivity of the MOS sensor to 2-CEES. Meanwhile, the sensor also showed good sensitivity, detection limits, response/recovery times, anti-interference, and stability, which further verified the feasibility of the improved scheme. Full article

(This article belongs to the Special Issue Advanced Sensors in Atomic Level)

► Show Figures

Figure 1

26 pages, 2058 KiB

Open AccessArticle

Web Real-Time Communications-Based Unmanned-Aerial-Vehicle-Borne Internet of Things and Stringent Time Sensitivity: A Case Study

by Agnieszka Chodorek and Robert Ryszard Chodorek

Sensors 2025, 25(2), 524; https://doi.org/10.3390/s25020524 (registering DOI) - 17 Jan 2025

Abstract

The currently observed development of time-sensitive applications also affects wireless communication with the IoT carried by UAVs. Although research on wireless low-latency networks has matured, there are still issues to solve at the transport layer. Since there is a general agreement that classical [...] Read more.

The currently observed development of time-sensitive applications also affects wireless communication with the IoT carried by UAVs. Although research on wireless low-latency networks has matured, there are still issues to solve at the transport layer. Since there is a general agreement that classical transport solutions are not able to achieve end-to-end delays in the single-digit millisecond range, in this paper, the use of WebRTC is proposed as a potential solution to this problem. This article examines UAV-borne WebRTC-based IoT in an outdoor environment. The results of field experiments conducted under various network conditions show that, in highly reliable networks, UAV and WebRTC-based IoT achieved stable end-to-end delays well below 10 ms during error-free air-to-ground transmissions, and below 10 ms in the immediate vicinity of the retransmitted packet. The significant advantage of the WebRTC data channel over the classic WebSocket is also demonstrated. Full article

(This article belongs to the Special Issue New Methods and Applications for UAVs)

► Show Figures

Figure 1

13 pages, 8085 KiB

Open AccessArticle

Feasibility Study on Additive Manufacturing of Feed Horn Operating in D-Band

by Giuseppe Addamo, Lorenzo Scalcinati, Mario Zannoni, Oscar Antonio Peverini and Flaviana Calignano

Sensors 2025, 25(2), 523; https://doi.org/10.3390/s25020523 (registering DOI) - 17 Jan 2025

Abstract

This paper presents the outcomes of a feasibility study on the manufacturing of D-band horn antennas through the Low Power Bed Fusion process. Different prototypes have been realized and tested, showing nice results in terms of the co-polarization component. On the other hand, [...] Read more.

This paper presents the outcomes of a feasibility study on the manufacturing of D-band horn antennas through the Low Power Bed Fusion process. Different prototypes have been realized and tested, showing nice results in terms of the co-polarization component. On the other hand, a spurious cross-polarization component is present in the radiation pattern even in the principal planes, limiting the device to single-polarization applications. A mechanical study of the realized devices has been conducted on the horn internal channel to understand the reasons for this issue, and, subsequently, to address where to improve the manufacturing process. Full article

(This article belongs to the Special Issue Applications of Manufacturing and Measurement Sensors: 2nd Edition)

► Show Figures

Figure 1

22 pages, 6644 KiB

Open AccessArticle

A Transformer Encoder Approach for Localization Reconstruction During GPS Outages from an IMU and GPS-Based Sensor

by Kévin Cédric Guyard, Jonathan Bertolaccini, Stéphane Montavon and Michel Deriaz

Sensors 2025, 25(2), 522; https://doi.org/10.3390/s25020522 (registering DOI) - 17 Jan 2025

Abstract

Accurate localization is crucial for numerous applications. While several methods exist for outdoor localization, typically relying on GPS signals, these approaches become unreliable in environments subject to a weak GPS signal or GPS outage. Many researchers have attempted to address this limitation, primarily [...] Read more.

Accurate localization is crucial for numerous applications. While several methods exist for outdoor localization, typically relying on GPS signals, these approaches become unreliable in environments subject to a weak GPS signal or GPS outage. Many researchers have attempted to address this limitation, primarily focusing on real-time solutions. However, for applications that do not require real-time localization, these methods remain suboptimal. This paper presents a novel Transformer-based bidirectional encoder approach to address, in postprocessing, the localization challenges during GPS weak signal phases or GPS outages. Our method predicts the velocity during periods of weak or lost GPS signals and calculates the position through bidirectional velocity integration. Additionally, it incorporates position interpolation to ensure smooth transitions between active GPS and GPS outage phases. Applied to a dataset tracking horse positions—which features velocities up to 10 times those of pedestrians and higher acceleration—our approach achieved an average trajectory error below 3 m, while maintaining stable relative distance errors regardless of the GPS outage duration. Full article

(This article belongs to the Special Issue Advancing Sensor Technology with Artificial Intelligence: Innovations and Applications)

► Show Figures

Figure 1

12 pages, 3943 KiB

Open AccessArticle

Influence of Initial Gap, Voltage, and Additives on Zinc Microcolumn Morphology by Local Electrochemical Deposition

by Yi Liu and Fuliang Wang

Sensors 2025, 25(2), 521; https://doi.org/10.3390/s25020521 (registering DOI) - 17 Jan 2025

Abstract

Local electrochemical deposition (LECD) is an innovative additive manufacturing technology capable of achieving precise deposition of metallic microstructures. This study delves into the ramifications of pivotal operational parameters—namely, the initial electrode gap, deposition voltage, and additive concentration—on the morphology of zinc microcolumns fabricated [...] Read more.

Local electrochemical deposition (LECD) is an innovative additive manufacturing technology capable of achieving precise deposition of metallic microstructures. This study delves into the ramifications of pivotal operational parameters—namely, the initial electrode gap, deposition voltage, and additive concentration—on the morphology of zinc microcolumns fabricated through LECD. A holistic approach integrating experimental methodologies with finite element simulations was adopted to scrutinize the influence of these variables on the microcolumns’ dimensions, surface morphology, and structural integrity. The findings reveal that augmenting the initial electrode gap results in microcolumns with larger diameters. Conversely, the deposition voltage primarily modulates the formation rate without exerting a notable impact on the columns’ dimensional attributes. The incorporation of additives enhances surface smoothness and diminishes column diameters; however, an overabundance of additives adversely affects the overall microstructure. Optimal parameters for the production of high-quality zinc microcolumns were determined to be a deposition voltage of 3.4 V and an electrode gap of 10 μm. These discoveries contribute pivotal insights for the refinement of LECD processes, with particular relevance to biomedical applications, such as the development of zinc-based bioabsorbable materials for orthopedic implants and cardiovascular devices. Full article

(This article belongs to the Section Chemical Sensors)

► Show Figures

Figure 1

16 pages, 5402 KiB

Open AccessArticle

Research on Sensitivity Improvement Methods for RTD Fluxgates Based on Feedback-Driven Stochastic Resonance with PSO

by Rui Wang, Na Pang, Haibo Guo, Xu Hu, Guo Li and Fei Li

Sensors 2025, 25(2), 520; https://doi.org/10.3390/s25020520 (registering DOI) - 17 Jan 2025

Abstract

With the wide application of Residence Time Difference (RTD) fluxgate sensors in Unmanned Aerial Vehicle (UAV) aeromagnetic measurements, the requirements for their measurement accuracy are increasing. The core characteristics of the RTD fluxgate sensor limit its sensitivity; the high-permeability soft magnetic core is [...] Read more.

With the wide application of Residence Time Difference (RTD) fluxgate sensors in Unmanned Aerial Vehicle (UAV) aeromagnetic measurements, the requirements for their measurement accuracy are increasing. The core characteristics of the RTD fluxgate sensor limit its sensitivity; the high-permeability soft magnetic core is especially easily interfered with by the input noise. In this paper, based on the study of the excitation signal and input noise characteristics, the stochastic resonance is proposed to be realized by adding feedback by taking advantage of the high hysteresis loop rectangular ratio, low coercivity and bistability characteristics of the soft magnetic material core. Simulink is used to construct the sensor model of odd polynomial feedback control, and the Particle Swarm Optimization (PSO) algorithm is used to optimize the coefficients of the feedback function so that the sensor reaches a resonance state, thus reducing the noise interference and improving the sensitivity of the sensor. The simulation results show that optimizing the odd polynomial feedback coefficients with PSO enables the sensor to reach a resonance state, improving sensitivity by at least 23.5%, effectively enhancing sensor performance and laying a foundation for advancements in UAV aeromagnetic measurement technology. Full article

(This article belongs to the Special Issue Magnetic and Electromagnetic Sensors for Engineered Systems and Technologies)

► Show Figures

Figure 1

22 pages, 15460 KiB

Open AccessArticle

The Application of an Intelligent Agaricus bisporus-Harvesting Device Based on FES-YOLOv5s

by Hao Ma, Yulong Ding, Hongwei Cui, Jiangtao Ji, Xin Jin, Tianhang Ding and Jiaoling Wang

Sensors 2025, 25(2), 519; https://doi.org/10.3390/s25020519 (registering DOI) - 17 Jan 2025

Abstract

To address several challenges, including low efficiency, significant damage, and high costs, associated with the manual harvesting of Agaricus bisporus, in this study, a machine vision-based intelligent harvesting device was designed according to its agronomic characteristics and morphological features. This device mainly [...] Read more.

To address several challenges, including low efficiency, significant damage, and high costs, associated with the manual harvesting of Agaricus bisporus, in this study, a machine vision-based intelligent harvesting device was designed according to its agronomic characteristics and morphological features. This device mainly comprised a frame, camera, truss-type robotic arm, flexible manipulator, and control system. The FES-YOLOv5s deep learning target detection model was used to accurately identify and locate Agaricus bisporus. The harvesting control system, using a Jetson Orin Nano as the main controller, adopted an S-curve acceleration and deceleration motor control algorithm. This algorithm controlled the robotic arm and the flexible manipulator to harvest Agaricus bisporus based on the identification and positioning results. To confirm the impact of vibration on the harvesting process, a stepper motor drive test was conducted using both trapezoidal and S-curve acceleration and deceleration motor control algorithms. The test results showed that the S-curve acceleration and deceleration motor control algorithm exhibited excellent performance in vibration reduction and repeat positioning accuracy. The recognition efficiency and harvesting effectiveness of the intelligent harvesting device were tested using recognition accuracy, harvesting success rate, and damage rate as evaluation metrics. The results showed that the Agaricus bisporus recognition algorithm achieved an average recognition accuracy of 96.72%, with an average missed detection rate of 2.13% and a false detection rate of 1.72%. The harvesting success rate of the intelligent harvesting device was 94.95%, with an average damage rate of 2.67% and an average harvesting yield rate of 87.38%. These results meet the requirements for the intelligent harvesting of Agaricus bisporus and provide insight into the development of intelligent harvesting robots in the industrial production of Agaricus bisporus. Full article

(This article belongs to the Section Sensing and Imaging)

► Show Figures

Figure 1

Journal Description

Sensors

Latest Articles

Journal Menu

Journal Browser

Highly Accessed Articles

Latest Books

E-Mail Alert

News

Topics

Conferences

Special Issues

Topical Collections

Further Information

Guidelines

MDPI Initiatives

Follow MDPI