Search Results (5,290)

Background: Electroencephalographic (EEG) functional connectivity analysis requires multiple signal-processing, source-modelling, and statistical steps that can limit its adoption in clinician-led randomised controlled trials (RCTs). NeuroStat was developed as a prototype research tool to integrate this workflow; formal usability validation with clinician end-users has not yet been conducted. Methods: NeuroStat is an open-source Python/PyQt6 desktop application that integrates automated artefact removal (a Generalised Eigenvalue Decomposition for Artefact Identification [GEDAI] pathway and a traditional Artefact Subspace Reconstruction (ASR)/Independent Component Analysis (ICA)/ICLabel pathway), boundary element model (BEM) source localisation using the Desikan–Killiany atlas (68 cortical regions), Phase Lag Index (PLI) connectivity estimation across five canonical frequency bands, and RCT-oriented statistical analysis. Evaluation separated sensor-space and source-space claims: a sensor-level simulation (repeated across five independent random seeds) tested preprocessing robustness, a repeated source-space simulation tested recovery of a known cortical parcel-pair contrast after forward projection and inverse reconstruction, a PhysioNet benchmark tested posterior Desikan–Killiany alpha PLI in 20 healthy adults, and an illustrative application to 20 sessions from a published chiropractic RCT demonstrated real-world workflow applicability. Results: In the sensor-level simulation benchmark, the Traditional pathway achieved a mean absolute error of $0.168\pm 0.017$ PLI units and root mean squared error of $0.219\pm 0.045$ (mean ± SD across five independent random seeds) across all artefact conditions. In the source-space simulation, reconstructed alpha PLI for the known bilateral lateral-occipital parcel pair exceeded anterior control edges across 60 repeated condition runs (mean known-control difference = 0.105 PLI units, 95% CI 0.096–0.114; $t\left(59\right)=22.61$, $p<0.001$). In the PhysioNet source-space benchmark, posterior Desikan–Killiany alpha PLI was higher during eyes-closed than eyes-open rest (Cohen’s $d=0.85$, $p=0.001$; 16/20 subjects showing the expected direction) after ICLabel-enabled preprocessing. In the pilot RCT application, all 20 sessions completed processing without manual intervention, with default-mode network alpha PLI showing a pre-to-post change of +0.071 in the intervention group versus +0.015 in the active control group. Conclusions: NeuroStat integrates preprocessing, source-space construction, connectivity estimation, and statistical reporting within a parameter-logged desktop workflow for EEG functional connectivity studies. Current evidence supports initial technical feasibility, sensor-level preprocessing robustness for one pathway in controlled simulations, source-space recovery of a known parcel-level contrast, source-space sensitivity to an expected posterior alpha resting-state contrast, and error-free processing across 20 real RCT sessions in a pilot workflow demonstration. Formal usability testing, test–retest reliability analysis, participant-specific source-model validation, and clinical-population validation remain necessary before clinician-facing or trial-deployment claims can be made. Full article

Source-Seeking in unknown scalar fields is a fundamental problem in robotics with applications in environmental monitoring and disaster response. In this work, we present a source-seeking approach with non-reversing forward velocity regulation by fusing measurement data from multiple sensors within the Stochastic Extremum Seeking (SES) framework. Specifically, a device model with multiple sensors is first constructed, and then a velocity regulation scheme is designed by leveraging the boundedness of the hyperbolic tangent function and the non-negativity of the exponential function to guarantee strictly positive forward velocity. We then evaluate the algorithm both in simulation environments and on the real-world Two-Wheeled Differential Drive Robot platform. The experiments show that our approach not only ensures the forward velocity remains non-negative, aligning with the design expectation, but also accurately locates the source. This work provides new insights into the design of velocity regulation strategies within the SES framework. Full article

To address the challenge of hand intrusion detection in frozen tuna cutting operations where operators wear thermal-insulating gloves, this study proposes a hand detection method based on dual-domain background modeling with absolute accuracy constraints. To tackle issues arising from low-resolution infrared arrays, such as defective pixels, random noise, and complex low-temperature backgrounds, a data preprocessing pipeline integrating defective pixel correction, exponential moving average (EMA), and median filtering is developed. A dual-domain background suppression (DDBS) strategy, combining spatial-domain and temporal-domain models with sensor absolute accuracy constraints, is employed to extract hand foregrounds under complex thermal conditions. Temperature thresholding, connected-component analysis, and hole-filling are further applied to effectively separate hands from frozen tuna. An experimental platform incorporating a Raspberry Pi 4B and an MLX90640 sensor was constructed, and a dataset comprising 1173 infrared frames was collected for validation purposes. Experimental results demonstrate that the proposed method achieves an accuracy of 94.12%, precision of 91.69%, recall of 97.55%, and F1-score of 94.53% for hand intrusion detection, with an average processing time of approximately 1.84 ms per frame. This provides a cost-effective, real-time solution for hand safety monitoring in frozen food processing operations. Full article

In precision livestock farming, 3D point clouds provide important data support for analyzing pig behavior and monitoring their health. However, due to environmental occlusions, limited sensor viewpoints, and mutual shielding between pigs, the acquired point clouds are often severely partial, which affects the accuracy of body shape modeling and behavior recognition. To address these challenges, this study constructed a pig pose point cloud dataset using multi-view depth camera acquisition and point cloud registration techniques. Based on this dataset, an improved point cloud completion model, IIR-PoinTr, is proposed to enhance the reconstruction of geometric and topological structures in pig bodies. By strengthening local geometric perception and high-dimensional feature representation, the model improves the reconstruction quality of partial pig point clouds and produces more structurally consistent pig body shapes. Experimental results show that, on the self-constructed pig posture dataset, the proposed method reduces Chamfer Distance (CD-L1) by 3.6%, CD-L2 by 6.9%, and Earth Mover’s Distance (EMD) by 2.0%, while improving the F-score by 5.4% compared with the baseline model. In single-view point cloud completion tasks, the method is capable of reconstructing geometrically consistent pig body structures and increases downstream classification accuracy by 34.9%. These results indicate that the proposed method can improve the reconstruction quality of partial pig point clouds and provide preliminary technical support for posture analysis under occlusion. Full article

In modern power systems such as new energy generation and smart grids, inverters serve as core equipment for electrical energy conversion and transmission. Their operational reliability directly impacts system power supply quality and safety stability. Currently, research on inverter fault diagnosis technology primarily focuses on linear load conditions, with diagnostic method design and validation based on linear load characteristics. However, with the rapid advancement of power electronics technology, power electronic loads such as variable frequency drives, charging stations, and distributed power sources are increasingly prevalent in power systems. These loads exhibit nonlinear and time-varying characteristics under complex operating conditions, leading to a growing variety of inverter faults with significantly diversified and complex fault signatures. Traditional diagnostic methods fail to adapt to the unique characteristics of power electronic loads, making it difficult to accurately identify various faults. Consequently, they no longer meet the diagnostic demands of practical engineering scenarios. In addition, current diagnostic methods for open-circuit power transistors, intermittent faults, and sensor faults often employ different approaches, which consume significant controller resources and are prone to mutual interference, leading to false triggers. This paper takes a three-phase four-wire inverter as the research subject. Targeting the challenge of fault diagnosis under power electronic load conditions, it proposes a comprehensive diagnostic method capable of simultaneously diagnosing power switch open circuits, intermittent faults, and current sensor faults. First, the characteristics of various faults are analyzed. Subsequently, fault diagnosis variables are constructed using the actual arm voltage of the inverter and the ideal arm voltage. Logical rules for each type of fault are established, and diagnosis is performed through fuzzy logic inference. Finally, experiments validated the effectiveness of this fault diagnosis scheme, with open-circuit faults detected in less than 2 ms, intermittent faults in less than 0.5 ms, and sensor faults in less than 3 ms. Full article

Metal oxide nanomaterials tailored at the nanoscale are opening new avenues for advanced electroanalytical sensing devices with enhanced properties, including improved electrocatalytic activity. In this work, a novel Mn₂Mo₃O₈/MnO-MWCNT nanocomposite was employed to modify a screen-printed carbon electrode, enabling the fabrication of an amperometric sensor for H₂O₂ operating at relatively low applied potential due to the catalytic activity of the nanocomposite. Further functionalization of this nanostructured surface with glucose oxidase allowed the construction of an electrochemical glucose biosensor, where the Mn₂Mo₃O₈/MnO-MWCNT material acted as an efficient electrocatalyst for hydrogen peroxide detection. The H₂O₂ sensor exhibited a linear response from 0.06 mM to 3.00 mM, with a sensitivity of (2.22 ± 0.02) µA mM⁻¹ and a detection limit of 22 µM. The glucose biosensor showed a linear response in the range from 0.10 mM to 18.9 mM glucose, with a sensitivity of (0.345 ± 0.005) µA mM⁻¹, and a detection limit of 29 µM. The biosensor displayed excellent selectivity and high stability and was successfully applied to the determination of glucose in lactose-free skimmed milk. Full article

Edge-side acoustic monitoring enables animal sound recognition in remote environments, but microcontroller deployment remains constrained by feature extraction, numerical consistency, memory, latency, and energy consumption. This study presents a sensor-based tiny machine learning (TinyML) acoustic monitoring system on an Arduino Nano 33 BLE Sense Rev2 platform, integrating onboard pulse-density modulation (PDM) microphone acquisition, Mel-frequency cepstral coefficient (MFCC) feature extraction, deployment-side standardization, 8-bit integer (INT8) neural-network inference, and edge-side decision output. To reduce training-to-deployment feature drift, consistent frame parameters, mirrored C++ feature operators, and exported standardization parameters are used to align personal-computer-side and microcontroller-side feature representations. A source-isolated seven-class protocol was constructed for six target animal classes and one compound background-noise class. In the single-run baseline comparison, the proposed multilayer perceptron achieved 98.28% test accuracy and 97.21% test macro-F1, while the ten-seed stability analysis yielded 98.64% ± 0.26% test accuracy and 97.87% ± 0.38% test macro-F1. The deployed INT8 model occupied approximately 26.9 KB, with a post-window latency of about 303 ms. System-level input power was 0.783–0.825 W, corresponding to an estimated autonomy of 7.63–8.03 h under the reference battery setting. Full article

The increasing frequency of extreme weather events, particularly hailstorms, driven by climate change, poses growing threats to the resilience, environmental sustainability, and long-term performance of photovoltaic (PV) systems. This study evaluates the environmental impacts of a 12 kWp rooftop PV installation in Brescia, northern Italy, through a comparative Life Cycle Assessment (LCA) of three system configurations: a standard unprotected system (Scenario A), one equipped with a retractable polycarbonate hail-protection panel with automated weather-sensor activation (Scenario B), and one using thicker reinforced front-glass modules (Scenario C). The analysis follows a cradle-to-gate plus operational maintenance phase (30-year horizon, excluding end-of-life) system boundary and employs the ReCiPe 2016 Midpoint (H) methodology across 18 environmental impact categories. A novel integration of the Social Discount Rate (SDR) to the LCA framework—constituting a Discounted LCA (D-LCA)—incorporates both temporal discounting and risk dimensions into the environmental evaluation. A structured PESTEL-based risk taxonomy is applied to derive scenario-specific SDRs, with the Environmental risk category as the key differentiator between configurations. The static LCA identifies Scenario A as the lowest-impact option, while the D-LCA framework reverses this ranking: Scenario C achieves the highest Net Present Value of Emissions, followed by Scenario A. A negative NPV-E for Scenario B reflects the temporal cost of a large, front-loaded construction debt rather than absolute environmental harm. D-LCA framework should be interpreted as a complement to the full 18-category static LCIA profile, not a replacement. These results demonstrate that risk-informed D-LCA provides a more policy-relevant environmental sustainability assessment than static LCA for long-lived energy infrastructure subject to climate-driven operational risks. Full article

The dimensional accuracy of the bearing inner ring is critical for the operational performance and reliability of high-end equipment. However, nonlinear deformation of the measurement mechanism caused by temperature variations and temperature drift of the sensor significantly affect the measurement accuracy. In this study, a novel online measurement system for bearing inner diameter was designed, which integrates a two-degree-of-freedom motion mechanism and an adaptive elastic measurement probe. To compensate for the measurement errors caused by temperature effects in the proposed system, an intelligent compensation method based on a CNN-LSTM–Attention hybrid model was proposed. The raw sensor signals and ambient temperature were used as the model inputs, and an end-to-end nonlinear mapping relationship for the actual bearing inner diameter deviation was established without the need to construct complex explicit physical equations. The experimental results show that, within the investigated temperature interval of 11–21 °C, the proposed method controls the measurement error within 1.87 μm, thereby satisfying the dimensional measurement requirement for P4-grade bearings with a tolerance of 0 to −4 μm. Full article

Lane-changing and overtaking constitute a typical complex driving manoeuvre for intelligent vehicles operating on structured roads; this task demands that the vehicle not only plan a safe and smooth lane-change trajectory but also requires the control system to maintain high tracking accuracy and lateral stability. Addressing the challenges of real-time path planning and stable tracking control inherent in lane-changing and overtaking scenarios, this paper proposes a trajectory planning and control method that integrates an improved artificial potential field (APF) approach with a lateral–longitudinal cooperative controller. Regarding path planning, the proposed method constructs attractive and repulsive fields based on the APF framework, while introducing virtual target points, elliptical obstacle models, and velocity-dependent repulsive fields to mitigate the risk of local minima and enhance dynamic obstacle avoidance capabilities. To ensure trajectory continuity and trackability, a fifth-order polynomial is employed to smooth the planned path. Regarding control, the method utilises a Linear Quadratic Regulator (LQR)—optimised via a genetic algorithm—for lateral control; this is coupled with a dual-PID longitudinal controller that generates throttle and braking commands based on vehicle speed errors, thereby establishing a cooperative lateral–longitudinal tracking control strategy. The proposed method is validated using a CarSim–MATLAB/Simulink co-simulation platform. Simulation results demonstrate that the proposed method significantly improves trajectory-tracking accuracy and vehicle stability during lane-changing and overtaking manoeuvres. In a single lane-change scenario, the maximum lateral error is reduced from approximately 0.62 m to 0.22 m, and the heading angle error decreases from about 0.058 rad to 0.01 rad; in a continuous lane-changing scenario, the maximum lateral error drops from approximately 0.30 m to 0.04 m, while the heading angle error falls from about 0.016 rad to 0.005 rad. Furthermore, the yaw rate, sideslip angle, and lateral acceleration are reduced by 39.1%, 22.2%, and 28.9%, respectively. These results confirm that, under the specified simulation conditions, the proposed method exhibits superior tracking performance and stability. Future research could further explore more complex driving scenarios, such as curved roads, multi-vehicle interactions, sensor uncertainties, actuator delays, and real-vehicle field experiments. Full article

To address the limitations of vision-only defect detection, image–semantic misalignment, and spatial-logic conflicts in complex substation inspection scenarios, this paper proposes a camera-sensor-based multimodal defect sensing framework with cross-modal feature mapping for substation equipment monitoring. The proposed framework integrates field inspection images acquired by camera sensors, defect textual descriptions, and equipment topology knowledge and establishes a unified domain-adaptive pre-training–bidirectional cross-modal mapping–hierarchical reasoning workflow. First, a Contrastive Language–Image Pre-training (CLIP)-based domain-adaptive pre-training strategy is developed to enhance the representation of equipment categories, defect attributes, and inspection-scene semantics. Second, a bidirectional cross-modal feature mapping network is constructed to model fine-grained interactions between candidate visual regions and textual semantics, where uncertainty-aware fusion and prototype constraints are introduced to improve semantic alignment and defect discrimination. Third, a hierarchical neuro-symbolic reasoning module incorporates equipment topology and spatial rules for posterior verification, logical consistency checking, and false-positive suppression. Experiments on a substation inspection image dataset demonstrate that the proposed method achieves 90.8% mAP@0.5, 68.7% mAP@0.5:0.95, and 89.4% F1-score, outperforming mainstream and recent detection models. Full article

Industrial control systems (ICSs) are cyber–physical environments in which physical process data and network communication data are generated simultaneously. Existing studies have mainly focused on either sensor-based or network-based anomaly detection, making it difficult to capture diverse attack indicators and motivating the use of multimodal methods that can leverage complementary information from both modalities. In this paper, we propose an unsupervised multimodal anomaly detection framework for ICSs that jointly uses sensor and network modalities. For each modality, autoencoder-based single-modality models are trained in an unsupervised manner, and their anomaly scores and latent feature vectors are extracted. These outputs are temporally aligned to construct a time-aligned multimodal table, which is then used to implement and compare two fusion strategies: anomaly score fusion and latent feature fusion. In latent feature fusion, aligned modality-specific latent features are combined with canonical correlation analysis (CCA)-derived cross-modal correlation features. The experimental results showed that latent feature fusion achieved stable performance across multiple sensor–network encoder combinations. In particular, the gated recurrent unit–convolutional neural network (GRU–CNN) combination achieved the best F1-score of 0.9166 and ROC-AUC of 0.9795. In addition, the complementarity analysis showed that latent feature fusion recovered some missed detections by integrating complementary sensor and network evidence. These results demonstrate that latent feature fusion is an effective multimodal strategy for ICS anomaly detection. Full article

As tunnel engineering continues to advance toward deeper, longer, and more complex projects, the risks encountered during the construction phase have evolved into a combination of various disaster types and the accumulation of multiple contributing factors. Traditional empirical and semi-empirical risk management methods are increasingly revealing shortcomings in terms of timeliness, accuracy, and the ability to process multi-source data. In recent years, driven by advancements in computing power and sensor technology, artificial intelligence algorithms (AI algorithms) such as machine learning and deep learning have been rapidly adopted in tunnel construction risk management. This paper retrieved relevant literature from the Web of Science database covering the period from 2010 to 2025. After rigorous screening, 96 highly relevant papers were selected for bibliometric analysis. This paper systematically reviews research progress from two perspectives: algorithmic models and engineering applications. The review indicates that, in terms of algorithmic models, traditional machine learning, convolutional neural network, recurrent neural network, generative adversarial network, Transformer, and graph neural network constitute a multi-level technical framework encompassing feature representation, risk perception, and intelligent decision-making. In terms of applications, AI algorithms have been widely integrated into typical scenarios such as geological hazard identification and prediction, surrounding rock stability and deformation prediction, rock burst assessment and early warning, lining defect detection and structural safety assessment, construction-induced ground settlement prediction, and tunnel gas and fire hazard prediction, significantly enhancing risk identification and early warning capabilities. However, several challenges remain, including the scarcity of high-quality datasets, the prevalence of noisy, incomplete, and heterogeneous monitoring data, insufficient coupling between model interpretability and engineering mechanisms, limited cross-project transferability, and the lack of integrated management systems for multi-hazard lifecycle control. Based on this, this paper proposes future research directions in areas such as data infrastructure development, integration of mechanism constraints, and multi-hazard collaborative modeling, aiming to provide guidance for the further development of intelligent risk management in tunnel construction. Full article

Autonomous navigation of unmanned surface vehicles (USVs) in complex obstacle scenarios remains challenging due to redundant perception inputs, unstable value estimation, and inefficient policy convergence. To address these problems, this paper proposes LDA-D3QN, an improved deep reinforcement learning method for USV autonomous navigation. The proposed method constructs a compact navigation state representation by combining target-related information with local obstacle features, allowing the agent to retain key decision-making information while reducing unnecessary environmental redundancy. Based on this representation, an enhanced value-learning framework is developed to improve the stability of navigation decisions in cluttered environments. Moreover, a reward-guided and staged training strategy is introduced to help the agent gradually adapt to increasingly complex navigation tasks. The proposed method was evaluated on a Unity–ROS–MATLAB integrated simulation platform. Experimental results show that LDA-D3QN achieves superior overall navigation performance compared with several representative reinforcement learning algorithms. Specifically, the proposed method achieves a final training success rate of 91.4%, outperforming PPO (82.3%), Dueling DQN (78.5%), Double DQN (79.8%), and Rainbow DQN (86.5%). Additional tests in complex multi-obstacle and multi-target scenarios further demonstrate that the learned policy can generate safe, stable, and effective navigation behaviors. Preliminary validation using real-USV sensor data also confirms the feasibility of the LiDAR and GPS data processing procedures, providing a basis for future closed-loop autonomous navigation experiments and multi-sensor fusion deployment. Full article

Zero-shot remote sensing is attractive for scene classification because new regions, sensors, and label taxonomies often appear before sufficient annotated data are available for supervised adaptation. We present OATS-RS, an inference-centric framework that keeps a remote sensing vision–language model (VLM) backbone frozen and improves zero-shot decisions through ontology-aware prompt construction, hierarchical and contrastive scoring, adaptive multi-view aggregation, unlabeled transductive refinement, ambiguity-aware local re-ranking, and selective prediction. The method targets the common remote sensing regime in which neighboring classes such as annual crop, permanent crop, forest, pasture, herbaceous vegetation, river, and sea or lake overlap strongly in red–green–blue (RGB) appearance, meaning that they require more than a single class-name prompt. On the supplied final EuroSAT RGB evaluation with a GeoRSCLIP Contrastive Language–Image Pre-training (CLIP)-family Vision Transformer Base with 32 × 32-pixel patches (ViT-B-32) backbone, the complete pipeline obtains top-1 accuracy of 0.522, balanced accuracy of 0.522, macro-averaged F1 score (macro-F1) of 0.535, and top-3 accuracy of 0.887. The strongest classes are industrial area, residential area, river, highway, and pasture, whereas the weakest classes remain herbaceous vegetation and several fine-grained vegetation categories. Selective prediction increases accepted-example accuracy to 0.538 at 0.934 coverage, but the expected calibration error (ECE) remains high at 0.384. These results support a qualified conclusion: ontology-guided zero-shot inference can already recover useful semantic shortlists for structured remote-sensing scenes, but fine-grained natural-class disambiguation, calibrated confidence, multi-dataset transfer, component-level ablations, and measured runtime remain essential before dependable deployment claims can be made. Full article