Sensors

16 pages, 3523 KB

Open AccessArticle

Dynamical Artifacts in Knitted Resistive Strain Sensors: Effects of Conductive Yarns, Knitting Structures, and Loading Rates

by Alexander Oks Junior, Alexander Okss, Alexei Katashev and Uģis Briedis

Sensors 2026, 26(6), 2010; https://doi.org/10.3390/s26062010 - 23 Mar 2026

Viewed by 467

Abstract

This study investigates the dynamic artifacts (DAs) in knitted resistive strain sensors (KRSS) subjected to various deformation types, including stair-wise, trapezoidal, and triangle-type deformations. The presence of DAs, characterized by sharp peak-wise increases in resistance followed by a gradual decline, was observed across [...] Read more.

This study investigates the dynamic artifacts (DAs) in knitted resistive strain sensors (KRSS) subjected to various deformation types, including stair-wise, trapezoidal, and triangle-type deformations. The presence of DAs, characterized by sharp peak-wise increases in resistance followed by a gradual decline, was observed across all KRSS samples. The amplitude of DA peaks increased with higher deformation velocities within the investigated range of 2.6–40 cm/s. The study also identified the temporal offset between resistance and deformation during linear deformation, suggesting a complex mechanism underlying DAs. The results demonstrate that DAs are most prominent in stepwise and trapezoidal deformations, while continuous deformations like triangle-type loading partially mask these artifacts. The resistance signals were recorded at a sampling rate of 150 Hz, with temporal desynchronization between recorded parameters not exceeding 6.7 ms, enabling the observation of dynamic effects. Manifestation of DAs in KRSS degrades the metrological characteristics of KRSS and cannot be ignored. This paper provides insights into the relationship between KRSS structure, deformation velocity, and DA behavior, and provides an experimental basis for future compensation approaches to mitigate the impact of DAs on measurement accuracy. Full article

(This article belongs to the Section Wearables)

► Show Figures

Figure 1

43 pages, 6083 KB

Open AccessArticle

An Unscented Kalman Filter Based on the Adams–Bashforth Method with Applications to the State Estimation of Osprey-Type Drones Composed of Tiltable Rotor Mechanisms

by Keigo Watanabe, Soma Takeda and Isaku Nagai

Sensors 2026, 26(6), 2009; https://doi.org/10.3390/s26062009 - 23 Mar 2026

Viewed by 442

Abstract

In the state estimation problem for nonlinear systems, the Unscented Kalman Filter (UKF) has gained attention as an algorithm capable of accurate state estimation based on high-fidelity discretization for strongly nonlinear systems. Furthermore, for applying the UKF to continuous-time state–space models, a method [...] Read more.

In the state estimation problem for nonlinear systems, the Unscented Kalman Filter (UKF) has gained attention as an algorithm capable of accurate state estimation based on high-fidelity discretization for strongly nonlinear systems. Furthermore, for applying the UKF to continuous-time state–space models, a method employing the Runge–Kutta method in the time-update equation for sigma points has already been proposed to achieve high-precision state estimation. While this method uses high-order numerical approximations, the associated decrease in computational efficiency due to processing time becomes problematic. It is thus unsuitable for the state estimation of relatively fast-moving objects, such as autonomous vehicles and drones, which require high sampling frequencies. In this study, to reduce computational load while achieving relatively high estimation accuracy, we newly apply the Adams–Bashforth method to the UKF algorithm. The effectiveness of the proposed method is demonstrated by first explaining a low-dimensional model’s state estimation problem, followed by a comparison of estimation accuracy and computation time in state estimation simulations for the UAV model of an Osprey-type drone. Full article

(This article belongs to the Special Issue UAVs as Mobile Sensing Platforms: Advances, Innovations, and Emerging Applications)

► Show Figures

Figure 1

36 pages, 5099 KB

Open AccessArticle

DML–LLM Hybrid Architecture for Fault Detection and Diagnosis in Sensor-Rich Industrial Systems

by Yu-Shu Hu, Saman Marandi and Mohammad Modarres

Sensors 2026, 26(6), 2008; https://doi.org/10.3390/s26062008 - 23 Mar 2026

Viewed by 734

Abstract

Fault Detection and Diagnosis (FDD) in complex industrial systems requires methods that can handle uncertain operating conditions, soft thresholds, evolving sensor behavior, and increasing volumes of heterogeneous data. Traditional model-based or rule-driven approaches offer interpretability but lack adaptability, while purely data-driven and Large [...] Read more.

Fault Detection and Diagnosis (FDD) in complex industrial systems requires methods that can handle uncertain operating conditions, soft thresholds, evolving sensor behavior, and increasing volumes of heterogeneous data. Traditional model-based or rule-driven approaches offer interpretability but lack adaptability, while purely data-driven and Large Language Model (LLM)-based methods often struggle with consistency, traceability, and causal grounding. Dynamic Master Logic (DML) provides a causal and temporal reasoning structure with fuzzy rules that capture gradual drift, soft limits, and asynchronous sensor signals while preserving traceability and deterministic evidence propagation. Building on this foundation, this paper presents a DML–LLM hybrid architecture that integrates targeted LLM inference to interpret unstructured information such as logs, notes, or retrieved documents under controlled prompts that maintain domain constraints. The combined system integrates Bayesian updating, deterministic routing, and semantic interpretation into a unified FDD pipeline. In a semiconductor manufacturing case study, the proposed framework reduced time to detection (TTD) from 7.4 h to 1.2 h and improved the F1 score from 0.59 to 0.83 when compared with conventional Statistical Process Control (SPC) and Fault Detection and Classification (FDC) workflows. Provenance completeness increased from 18% to 96%, while engineer triage time was reduced from 72 min to 18 min per event. These results demonstrate that the hybrid framework provides a scalable and explainable approach to anomaly detection and fault diagnosis in sensor-rich industrial environments. Full article

(This article belongs to the Special Issue Anomaly Detection and Fault Diagnosis in Sensor Networks)

► Show Figures

Figure 1

11 pages, 8590 KB

Open AccessArticle

Optical Caliper for Contactless Measurement of Plant Stem Diameter

by Naomi van der Kolk, Daan Boesten, Willem van Valenberg and Steven van den Berg

Sensors 2026, 26(6), 2007; https://doi.org/10.3390/s26062007 - 23 Mar 2026

Viewed by 491

Abstract

Precision greenhouse agriculture enhances plant health and crop yields by continuously monitoring key plant parameters. Stem diameter is such a parameter and is monitored to support decisions on plant care. However, traditional contact-based methods induce thigmomorphogenic effects that impact plant growth. Here, we [...] Read more.

Precision greenhouse agriculture enhances plant health and crop yields by continuously monitoring key plant parameters. Stem diameter is such a parameter and is monitored to support decisions on plant care. However, traditional contact-based methods induce thigmomorphogenic effects that impact plant growth. Here, we introduce the Optical Caliper (OC), a novel contactless device for precise, non-invasive stem diameter measurement. The OC operates by projecting a collimated light beam to cast a shadow of the stem onto a high-resolution image sensor. The shadow size is a measure for the stem diameter. Controlled laboratory tests show the OC offers an accuracy comparable to that of a Digital Caliper (DC). Field trials on irregular tomato and cucumber stems demonstrate a repeatability of 0.1–0.2 mm. The OC’s non-invasive design and high repeatability exceed the performance of a DC, making it particularly suited for accurately monitoring soft, variable plant structures. Bringing the advantage of avoiding thigmomophogenic effects and thus optimizing crop yield, the OC is a promising tool for high-throughput plant phenotyping and precision agriculture applications. Full article

(This article belongs to the Special Issue Optical Techniques and Sensors for Biological and Agricultural Applications)

► Show Figures

Figure 1

27 pages, 3012 KB

Open AccessArticle

Emergency Operation Scheme Generation for Urban Rail Transit Train Door Systems Using Retrieval-Augmented Large Language Models

by Lu Huang, Zhigang Liu, Chengcheng Yu, Tianliang Zhu and Bing Yan

Sensors 2026, 26(6), 2006; https://doi.org/10.3390/s26062006 - 23 Mar 2026

Viewed by 653

Abstract

Urban rail transit (URT) train-door failures are safety-critical and can cause cascading service disruptions, yet existing emergency operation schemes (EOSs) are often static, difficult to adapt to evolving fault patterns, and hard to verify against updated regulations. This study proposes a retrieval-augmented large [...] Read more.

Urban rail transit (URT) train-door failures are safety-critical and can cause cascading service disruptions, yet existing emergency operation schemes (EOSs) are often static, difficult to adapt to evolving fault patterns, and hard to verify against updated regulations. This study proposes a retrieval-augmented large language model (LLM) framework for executable and evidence-traceable EOS generation. Multi-source heterogeneous incident evidence (structured work orders, operational impact records, and unstructured maintenance/dispatch narratives) is normalized into a structured incident representation, and a hybrid retriever (dense + BM25) with cross-encoder reranking selects compact regulatory clauses and historical cases under a fixed context budget. The generator is fine-tuned with structured objectives to enforce schema compliance, role assignment, and citation grounding. Experiments on 776 passenger-door incidents from Shanghai URT (2019–2024) show that Hybrid + rerank achieves the best retrieval quality (Recall@5 = 0.78; Coverage@B = 0.71; FirstHit/B = 0.46). For generation, the full setting improves operational usability, reaching SchemaPass = 0.88, RoleAcc = 0.91, CiteCov = 0.73, and UsableAns = 0.83, compared with 0.15 UsableAns for a pure LLM baseline and 0.26 for prompting with RAG only. These results indicate that combining high-utility retrieval with structure- and citation-aware fine-tuning substantially improves the executability and verifiability of safety-critical operation schemes. Full article

(This article belongs to the Special Issue Advanced Sensing Technologies for Sustainable and Resilient Railway Infrastructures)

► Show Figures

Figure 1

23 pages, 11748 KB

Open AccessArticle

Polarization-Regularized Adversarial Pruning for Efficient Radio Frequency Fingerprint Identification on IoT Devices

by Caidan Zhao, Haoliang Jiang, Jinhui Yu, Zepeng Meng and Xuhao He

Sensors 2026, 26(6), 2005; https://doi.org/10.3390/s26062005 - 23 Mar 2026

Viewed by 464

Abstract

Radio frequency fingerprint identification (RFFI) based on physical-layer characteristics provides a reliable solution for secure authentication of Internet of Things (IoT) devices. Deep neural networks have demonstrated strong capability in improving RFFI performance; however, their high computational complexity and large parameter size pose [...] Read more.

Radio frequency fingerprint identification (RFFI) based on physical-layer characteristics provides a reliable solution for secure authentication of Internet of Things (IoT) devices. Deep neural networks have demonstrated strong capability in improving RFFI performance; however, their high computational complexity and large parameter size pose significant challenges for deployment on resource-constrained edge devices. In RFFI tasks, existing pruning methods often lack effective performance recovery strategies, which leads to noticeable degradation in identification accuracy after pruning. To address this issue, this paper proposes a pruning method based on adversarial learning and polarization regularization. Polarization regularization is applied to learnable soft masks to effectively distinguish channels to be pruned from those to be retained. In addition, an adversarial learning-based performance recovery strategy is introduced to align the output feature distributions between the baseline network and the pruning network, thereby improving identification accuracy after pruning. Experimental results on multiple RFFI datasets demonstrate that the proposed method can effectively prune ResNet18 and VGG16, achieving substantial reductions in model complexity with only minor losses in identification performance. Full article

(This article belongs to the Special Issue Advanced Robust Processing Techniques and Sensor Technologies for Complex Radio Frequency Signals)

► Show Figures

Figure 1

33 pages, 3657 KB

Open AccessReview

Electrochemical Biosensing Platforms for Rapid and Early Diagnosis of Crop Fungal and Viral Diseases

by Yuhong Zheng, Li Fu, Jiale Yang, Shansong Gao, Haobo Sun and Fan Zhang

Sensors 2026, 26(6), 2004; https://doi.org/10.3390/s26062004 - 23 Mar 2026

Viewed by 624

Abstract

Crop fungal and viral diseases cause annual economic losses exceeding USD 150 billion globally, demanding rapid, sensitive, and field-deployable diagnostic technologies. This review critically evaluates recent advances in electrochemical biosensing platforms for early crop pathogen detection, focusing on immunosensors, genosensors, aptasensors, and VOC-based [...] Read more.

Crop fungal and viral diseases cause annual economic losses exceeding USD 150 billion globally, demanding rapid, sensitive, and field-deployable diagnostic technologies. This review critically evaluates recent advances in electrochemical biosensing platforms for early crop pathogen detection, focusing on immunosensors, genosensors, aptasensors, and VOC-based systems. Reported analytical performances demonstrate ultralow detection capabilities, including 0.3 fg mL⁻¹ for viral coat proteins, 15 DNA copies for bacterial pathogens, 0.5 fg µL⁻¹ RNA detection for viroids, and nanomolar-level VOC sensing (35–62 nM), with response times ranging from 2 to 60 min. Comparative analysis reveals that genosensors and aptasensors generally achieve the lowest LODs due to nucleic acid amplification or high-affinity recognition, while immunosensors provide robust protein-level specificity validated against ELISA. Volatile organic compound (VOC) sensors enable non-invasive, pre-symptomatic monitoring but face specificity challenges. Despite strong laboratory performance, practical adoption is limited by matrix-derived electrochemical interference, environmental instability of biorecognition elements, workflow complexity, and insufficient standardization across studies. Emerging innovations, including magnetic bead enrichment, nanoporous and graphene-based electrodes, microfluidic integration, AI-assisted impedance interpretation, and biodegradable substrates, are progressively addressing these bottlenecks. This review emphasizes that successful field translation requires holistic workflow engineering, matrix-matched validation, and harmonized performance metrics rather than incremental sensitivity improvements alone. By integrating analytical chemistry, nanomaterials engineering, and agricultural decision-support frameworks, electrochemical biosensing platforms hold significant potential to enable decentralized, rapid, and sustainable crop disease management. Full article

(This article belongs to the Special Issue Electrochemical Biosensing Devices and Their Applications)

► Show Figures

Figure 1

19 pages, 7352 KB

Open AccessArticle

Track-to-Track Fusion for Cooperative Perception Using Collective Perception Messages

by Redge Melroy Castelino, Shrijal Pradhan and Axel Hahn

Sensors 2026, 26(6), 2003; https://doi.org/10.3390/s26062003 - 23 Mar 2026

Viewed by 490

Abstract

Vehicle-to-everything communication grants connected and automated road vehicles the opportunity to share their sensor information such as detected road objects for collective awareness. This paper compares various state fusion strategies within a high-level cooperative perception architecture, focusing on the fusion of object-level information [...] Read more.

Vehicle-to-everything communication grants connected and automated road vehicles the opportunity to share their sensor information such as detected road objects for collective awareness. This paper compares various state fusion strategies within a high-level cooperative perception architecture, focusing on the fusion of object-level information provided in standard Collective Perception Messages. This work compares five track-to-track fusion methods, namely Covariance Intersection, Inverse Covariance Intersection, Adapted Extended Kalman Filter, Adapted Unscented Kalman Filter and Information Matrix Fusion, using a simulation framework built with CARLA and Autoware. The methods are analyzed in a case study to assess their performance under different vehicle maneuvers and varying input information accuracy. The case study highlights trade-offs between fusion strategies and illustrate their behavior in asynchronous multi-agent scenarios. While the analysis is conducted in simulation, the architecture is designed to be extensible, and directions for future development are outlined, including the integration of classification and object confidence fusion modules. Full article

(This article belongs to the Special Issue Cooperative Perception and Control for Autonomous Vehicles)

► Show Figures

Figure 1

23 pages, 56439 KB

Open AccessArticle

Multipath Credibility Selection for Robust UWB Angle-of-Arrival Estimation in Narrow Underground Corridors

by Jianjia Li, Baoguo Yu, Songzuo Cui, Menghuan Yang, Jun Zhao, Runjia Su and Runze Tian

Sensors 2026, 26(6), 2002; https://doi.org/10.3390/s26062002 - 23 Mar 2026

Viewed by 503

Abstract

Waveguide-like propagation in elongated underground environments—utility corridors, logistics tunnels—generates dense multipath that can cause the earliest or strongest resolvable channel impulse response (CIR) component to originate from a specular reflection rather than the direct line-of-sight (LOS) path. In the single-anchor CIR-tap-based implementations common [...] Read more.

Waveguide-like propagation in elongated underground environments—utility corridors, logistics tunnels—generates dense multipath that can cause the earliest or strongest resolvable channel impulse response (CIR) component to originate from a specular reflection rather than the direct line-of-sight (LOS) path. In the single-anchor CIR-tap-based implementations common to practical ultra-wideband (UWB) systems, baseline estimators such as phase-difference-of-arrival (PDOA) and MUSIC rely on selecting a single dominant CIR component, producing large angle-of-arrival (AoA) errors whenever the selected path is a reflection. We propose a multipath credibility selection (MCS) AoA estimator, MCS-AoA, that does not require explicit LOS/NLOS classification. The algorithm scores each resolvable CIR component with four credibility factors—amplitude significance, time-of-flight (TOF) consistency, inter-baseline phase–geometry agreement, and cross-baseline coherence—and fuses retained candidates into a credibility-weighted spatial covariance matrix for 2D MUSIC search. Field experiments on a custom five-channel coherent UWB platform compare MCS-AoA against six baselines—PDOA, MUSIC, MVDR/Capon, TLS-ESPRIT, PwMUSIC, and DNN-AoA. In an underground corridor (5–40 m), MCS-AoA achieves an azimuth/elevation MAE of 1.00°/1.46°, outperforming all baselines (PDOA: 2.26°/2.49°; MUSIC: 1.76°/2.40°; next-best PwMUSIC: 1.44°/2.17°); in a logistics tunnel (5–80 m), it achieves a 1.19° overall azimuth MAE. Simulations corroborate these gains, with a 0.71° azimuth RMSE at 80 m (69.3% reduction over PDOA) and 86.6% of estimates falling within 1°. Full article

(This article belongs to the Section Navigation and Positioning)

► Show Figures

Figure 1

21 pages, 6048 KB

Open AccessArticle

Enhanced Multi-Scale Defect Detection in Steel Surfaces via Innovative Deep Learning Architecture

by Zhaoxuan Zhou and Yan Cao

Sensors 2026, 26(6), 2001; https://doi.org/10.3390/s26062001 - 23 Mar 2026

Viewed by 560

Abstract

Steel surface defects significantly impact product quality and safety in industrial settings. Traditional defect detection methods suffer from inefficiencies and limitations. This study introduces an innovative deep learning architecture, CTG-YOLO, designed to enhance multi-scale defect detection accuracy on steel surfaces. By integrating a [...] Read more.

Steel surface defects significantly impact product quality and safety in industrial settings. Traditional defect detection methods suffer from inefficiencies and limitations. This study introduces an innovative deep learning architecture, CTG-YOLO, designed to enhance multi-scale defect detection accuracy on steel surfaces. By integrating a CBY parallel network structure, a TFF-PANet neck network, and a GS-Head detection head, our model achieves superior feature extraction and fusion capabilities. Experimental results on the NEU-DET and GC10-DET datasets demonstrate significant improvements, with mean Average Precision (mAP) scores of 76.55% and 69.94%, respectively, outperforming the original YOLOv8s by 3.72% and 3.14%. This research provides a robust foundation for industrial defect detection applications. Full article

(This article belongs to the Section Fault Diagnosis & Sensors)

► Show Figures

Figure 1

12 pages, 2042 KB

Open AccessArticle

Performance Characterization and Optimization of a Miniaturized SERF Atomic Magnetometer via Tunable Laser Power

by Peng Shi, Chen Zuo, Qisong Li and Shulin Zhang

Sensors 2026, 26(6), 2000; https://doi.org/10.3390/s26062000 - 23 Mar 2026

Viewed by 1051

Abstract

Spin-exchange relaxation-free (SERF) atomic magnetometers have emerged as highly promising candidates for ultra-weak magnetic field detection, particularly in biomagnetic imaging, owing to their exceptional sensitivity, amenability to miniaturization, and near-room-temperature operation. While current miniaturized magnetometers typically employ laser chips with fixed optical power, [...] Read more.

Spin-exchange relaxation-free (SERF) atomic magnetometers have emerged as highly promising candidates for ultra-weak magnetic field detection, particularly in biomagnetic imaging, owing to their exceptional sensitivity, amenability to miniaturization, and near-room-temperature operation. While current miniaturized magnetometers typically employ laser chips with fixed optical power, the quantitative impact of laser power on critical performance metrics remains to be fully elucidated. This study systematically investigates the influence of laser power on sensitivity, bandwidth, and dynamic range by incorporating considerations of power broadening, saturation absorption, and noise constraints. A miniaturized probe, integrated with an actively controlled vertical-cavity surface-emitting laser (VCSEL), was developed for experimental validation. Theoretical and experimental results consistently demonstrate that as optical power increases, sensitivity exhibits a non-monotonic dependence, whereas both bandwidth and dynamic range manifest a monotonic upward trend, aligning well with theoretical simulations. The optimized sensor achieved a peak sensitivity of 16 fT/√Hz at 300 μW, while the bandwidth and dynamic range reached 230 Hz and ±5.4 nT at 500 μW, respectively. This work establishes a robust theoretical and experimental framework for the comprehensive performance optimization of laser-integrated miniaturized atomic magnetometers. Full article

(This article belongs to the Section Optical Sensors)

► Show Figures

Figure 1

45 pages, 10337 KB

Open AccessReview

Design, Implementation, and Advances in Indirect SERS Sensors for Biomedical and Human-Health-Related Analyte Detection

by North Pinkley, Uchhwas Banik, Nayeem Anam, Aastha Oza, Kevin J. Ledford and Bhavya Sharma

Sensors 2026, 26(6), 1999; https://doi.org/10.3390/s26061999 - 23 Mar 2026

Viewed by 815

Abstract

Novel, accurate molecular diagnostics are driving new advances across medicine, public health, and environmental monitoring. Surface-enhanced Raman spectroscopy (SERS) nanotags are powerful platforms for ultrasensitive, multiplexed, and quantitative detection of molecular targets. This review focuses on indirect sensing strategies, where SERS nanotags act [...] Read more.

Novel, accurate molecular diagnostics are driving new advances across medicine, public health, and environmental monitoring. Surface-enhanced Raman spectroscopy (SERS) nanotags are powerful platforms for ultrasensitive, multiplexed, and quantitative detection of molecular targets. This review focuses on indirect sensing strategies, where SERS nanotags act as signal transducers, resulting in enhanced and unique Raman spectra upon binding of target analytes (high specificity) and allowing for ultralow limits of detection. These indirect SERS sensors typically consist of a plasmonic core, a Raman reporter molecule, and a ligand that targets the analyte of interest. Each of these components contributes to the sensitivity, stability, and selectivity of the system. Rational design of SERS nanotags requires balancing enhancement efficiency with reproducibility, biocompatibility, and assay integration. The choice of reporter molecules, for instance, governs spectral uniqueness and enables multiplexed detection of multiple analytes within a single sample. Recent advances in artificial intelligence and machine learning are accelerating nanotag development by enabling predictive control over nanostructure geometry, composition, and optical response. SERS nanotags are increasingly being integrated into diagnostic formats, such as lateral flow assays and microfluidic devices, offering both qualitative and quantitative analysis at the point of care. This review provides an overview of key design principles, common strategies for nanostructure functionalization and stabilization, and emerging biosensing applications, serving as a practical guide for researchers seeking to design and implement SERS nanotags. Full article

(This article belongs to the Special Issue Spectral Sensing Techniques in Biological Detection and Analysis)

► Show Figures

Figure 1

23 pages, 1109 KB

Open AccessReview

Strategies for Class-Imbalanced Learning in Multi-Sensor Medical Imaging

by Da Zhou, Song Gao and Xinrui Huang

Sensors 2026, 26(6), 1998; https://doi.org/10.3390/s26061998 - 23 Mar 2026

Viewed by 634

Abstract

This narrative critical review addresses class imbalance in medical imaging, particularly within the context of multi-sensor and multi-modal environments, poses a critical challenge to developing reliable AI diagnostic systems. The integration of heterogeneous data from sources like CT, MRI, and PET presents a [...] Read more.

This narrative critical review addresses class imbalance in medical imaging, particularly within the context of multi-sensor and multi-modal environments, poses a critical challenge to developing reliable AI diagnostic systems. The integration of heterogeneous data from sources like CT, MRI, and PET presents a unique opportunity to address data scarcity for rare conditions through fusion techniques. This review provides a structured analysis of strategies to tackle class imbalance, categorizing them into data-centric (e.g., advanced resampling like SMOTE-ENC for mixed data types, GAN-based synthesis) and model-centric (e.g., loss function engineering, transfer learning, and ensemble methods) approaches. Crucially, we highlight how multi-sensor feature fusion and decision-level fusion paradigms can inherently enrich representations for minority classes, offering a powerful frontier beyond single-modality learning. We evaluate each method’s merits, clinical viability, and compliance considerations (e.g., FDA). Finally, we identify emerging trends where imbalance-aware learning synergizes with multi-sensor fusion frameworks, federated learning, and explainable AI, charting a roadmap toward robust, equitable, and clinically deployable diagnostic tools. Our quantitative synthesis shows that data-centric strategies can improve minority class recall by 12–35% in datasets with imbalance ratios (majority:minority) ≥10:1, while model-centric strategies achieve an average AUC improvement of 0.08–0.21 in multi-sensor medical imaging tasks with sample sizes ranging from 50 to 50,000. Full article

(This article belongs to the Special Issue Multi-sensor Fusion in Medical Imaging, Diagnosis and Therapy)

► Show Figures

Figure 1

17 pages, 561 KB

Open AccessArticle

Multimodal Shared Autonomy for Heavy-Load UAV Operations with Physics-Aware Cooperative Control

by Xu Gao, Jingfeng Wu, Yuchen Wang, Can Cao, Lihui Wang, Bowen Wang and Yimeng Zhang

Sensors 2026, 26(6), 1997; https://doi.org/10.3390/s26061997 - 23 Mar 2026

Viewed by 467

Abstract

Heavy-load unmanned aerial vehicles (UAVs) are increasingly being applied in logistics, infrastructure installation, and emergency response missions, where complex payload dynamics and unstructured environments pose significant challenges to safe and efficient operation. Conventional manual teleoperation interfaces, such as dual-joystick control, impose a high [...] Read more.

Heavy-load unmanned aerial vehicles (UAVs) are increasingly being applied in logistics, infrastructure installation, and emergency response missions, where complex payload dynamics and unstructured environments pose significant challenges to safe and efficient operation. Conventional manual teleoperation interfaces, such as dual-joystick control, impose a high cognitive workload and provide limited support for expressing high-level operator intent, while fully autonomous solutions remain difficult to deploy reliably under real-world uncertainty. To address these limitations, this paper proposes the Multimodal Fusion Cooperation Network (MFCN), an end-to-end shared autonomy framework that integrates speech commands, visual gestures, and haptic cues through cross-modal feature fusion to infer operator intent in real time. The fused intent representation is translated into dynamically feasible control commands by a cooperative control policy with embedded physics-aware constraints to suppress payload oscillations and ensure flight stability. Extensive semi-physical simulations and real-world experiments demonstrate that the MFCN significantly improves the task success rate, positioning accuracy, and payload stability while reducing the task completion time and operator cognitive workload compared with manual, unimodal, and heuristic multimodal baselines. Full article

(This article belongs to the Special Issue Advanced Sensors and AI Integration for Human–Robot Teaming)

► Show Figures

Figure 1

13 pages, 601 KB

Open AccessArticle

Wearable-Based Assessment of Cardiac Recovery After a Modified Bruce Test in Women with Breast Cancer: Role of Physical Activity and Treatment Duration

by Carlos Navarro-Martínez, Natalia Ferrer-Artero, Keven Santamaria-Guzman and José Pino-Ortega

Sensors 2026, 26(6), 1996; https://doi.org/10.3390/s26061996 - 23 Mar 2026

Viewed by 543

Abstract

Heart rate recovery (HRR) is an important indicator of cardiovascular autonomic function, yet evidence in women with breast cancer remains limited. This study aimed to analyze heart rate recovery during the first two minutes following a maximal exercise test and to examine its [...] Read more.

Heart rate recovery (HRR) is an important indicator of cardiovascular autonomic function, yet evidence in women with breast cancer remains limited. This study aimed to analyze heart rate recovery during the first two minutes following a maximal exercise test and to examine its association with age, weekly physical activity, and oncological treatment duration using wearable technology. A cross-sectional design was applied in 22 women with breast cancer enrolled in an oncological exercise program. Participants performed a maximal treadmill test using the Modified Bruce Protocol, after which the mean heart rate was recorded across eight 15 s recovery intervals using a wearable chest-strap heart rate sensor integrated with an inertial device (WIMU PRO). Results showed a progressive and significant decrease in heart rate during recovery, with the first statistically significant pairwise difference emerging at 45–60 s post-exercise compared to the initial recovery interval (p < 0.05), within the context of a continuous HR decline. Regression analysis identified weekly physical activity hours (β = −0.281, p = 0.013) and oncological treatment duration (β = −0.245, p = 0.038) as significant predictors of mean heart rate recovery, explaining 4.8% of the variance, while age was not significantly associated (β = 0.049, p = 0.622). In conclusion, a differentiated recovery pattern emerged at approximately 45–60 s post-exercise, with weekly physical activity and oncological treatment duration as determinants. These findings support the use of wearable-based monitoring to inform individualized exercise prescription in women with breast cancer. Full article

(This article belongs to the Special Issue Flexible Electronics and Wearables for Biomedical, Health and Sport Monitoring)

► Show Figures

Figure 1

32 pages, 31110 KB

Open AccessArticle

Explicit Features Versus Implicit Spatial Relations in Geomorphometry: A Comparative Analysis for DEM Error Correction in Complex Geomorphological Regions

by Shuyu Zhou, Mingli Xie, Nengpan Ju, Changyun Feng, Qinghua Lin and Zihao Shu

Sensors 2026, 26(6), 1995; https://doi.org/10.3390/s26061995 - 23 Mar 2026

Viewed by 475

Abstract

Global Digital Elevation Models (DEMs) exhibit systematic biases constrained by acquisition geometry and surface penetration. This study aims to evaluate whether the increasing complexity of geometric deep learning (e.g., Graph Neural Networks, GNNs) is justified by performance gains over established feature engineering paradigms [...] Read more.

Global Digital Elevation Models (DEMs) exhibit systematic biases constrained by acquisition geometry and surface penetration. This study aims to evaluate whether the increasing complexity of geometric deep learning (e.g., Graph Neural Networks, GNNs) is justified by performance gains over established feature engineering paradigms (e.g., XGBoost) under the constraints of sparse altimetry supervision. We established a rigorous comparative framework across four mainstream products—ALOS World 3D, Copernicus DEM, SRTM GL1, and TanDEM-X—using Sichuan Province, China, as a representative natural laboratory. Our results reveal a fundamental scale mismatch (where the ~485 m average spacing of sampled altimetry footprints dwarfs the local terrain resolution): despite their topological complexity, Hybrid GNN models fail to establish a statistically significant accuracy advantage over the systematically optimized XGBoost baseline, demonstrating RMSE parity. Mechanistically, we uncover a critical divergence in decision logic: XGBoost relies on a stable “Physics Skeleton” consistently dominated by deterministic features (terrain aspect and vegetation density), whereas GNNs exhibit severe “Attribution Stochasticity” (

ρ \approx

0.63–0.77). The GNN component acts as a residual-dependent latent feature learner rather than discovering universal topological laws. We conclude that for geospatial regression tasks relying on sparse supervision, “Physics Trumps Geometry.” A “Feature-First” paradigm that prioritizes robust, domain-knowledge-based physical descriptors outweighs the indeterminate complexity of “Black Box” architectures. This study underscores the imperative of prioritizing explanatory stability over marginal accuracy gains to foster trusted Geo-AI. Full article

(This article belongs to the Section Remote Sensors)

► Show Figures

Figure 1

30 pages, 2818 KB

Open AccessReview

Nondestructive Inspection of Water Pipes: A Review

by Rileigh Nowroski, Piervincenzo Rizzo, Liam Byrne and Adeline Ziegler

Sensors 2026, 26(6), 1994; https://doi.org/10.3390/s26061994 - 23 Mar 2026

Viewed by 636

Abstract

Pipe networks assure the transportation of primary commodities such as water, oil, and natural gas. Quantitative and early detection of defects avoids costly consequences. Due to low cost of water, high-profile accidents, and economic downturns, the research and development of nondestructive evaluation (NDE) [...] Read more.

Pipe networks assure the transportation of primary commodities such as water, oil, and natural gas. Quantitative and early detection of defects avoids costly consequences. Due to low cost of water, high-profile accidents, and economic downturns, the research and development of nondestructive evaluation (NDE) and structural health monitoring (SHM) technologies for freshwater mains and urban water networks have received less attention with respect to the gas and oil industries. Moreover, the technical challenges associated with the practical deployment of monitoring systems and the fact that most water pipelines are buried underground demand synergistic interaction across several disciplines, which may limit the transition from laboratory to real structures. This paper reviews the most prominent NDE/SHM technologies for freshwater pipes. The challenges that said infrastructures pose, as well as the methodologies that can be translated into SHM approaches, are highlighted. The scope of this review is to provide a holistic view of the physical principles, the success, and the technological challenges associated with the inspection and monitoring of freshwater pipelines. Full article

(This article belongs to the Special Issue IoT Sensors and Systems for the Maintenance of Large Civil and Industrial Infrastructures)

► Show Figures

Figure 1

20 pages, 2863 KB

Open AccessArticle

Particle Filtering-Based In-Flight Icing Detection for Unmanned Aerial Vehicles

by Toufik Souanef, Mohamed Tadjine, Nadjim Horri, Ilyes Chaabeni and Bilel Boulassel

Sensors 2026, 26(6), 1993; https://doi.org/10.3390/s26061993 - 23 Mar 2026

Viewed by 409

Abstract

Ice accretion poses a threat to fixed-wing aerial vehicles as it alters the wings’ shape and thus degrades the aerodynamic performance. In manned aircraft, the icing detection system assists the pilot and utilises dedicated sensors. However, in unmanned aerial vehicles (UAVs), onboard icing [...] Read more.

Ice accretion poses a threat to fixed-wing aerial vehicles as it alters the wings’ shape and thus degrades the aerodynamic performance. In manned aircraft, the icing detection system assists the pilot and utilises dedicated sensors. However, in unmanned aerial vehicles (UAVs), onboard icing detection can generally only be achieved using standard sensors in conjunction with dynamical models, because dedicated sensors are rarely available. In this paper, we propose two approaches based on the particle filter for both icing detection and accurate state and aerodynamic parameter estimation in the presence of icing, with different levels of severity. The first approach uses the observation likelihood for icing hypothesis testing with a complement of the Gaussian kernel to compute icing probability. The second approach uses a discrete jump approach based on a Bernoulli process and a subset of particles to test the icing hypothesis for faster icing detection by estimating changes in icing-related aerodynamic parameters. Using both approaches, the simulation results demonstrate improved estimation accuracy compared to an extended Kalman filter (EKF), under both moderate and severe icing conditions. With adequate tuning, the proposed approaches show potential for indirect icing detection in UAVs. They also enable the computation of icing severity and provide a more accurate and reliable estimate of the icing probability compared to the EKF. Full article

(This article belongs to the Special Issue Connected and Intelligent Sensors and Smart Systems for Improved Vehicle Autonomy, Efficiency and Resilience)

► Show Figures

Figure 1

29 pages, 7304 KB

Open AccessReview

Enhanced Lateral Resolution in Acoustic Imaging: From High- to Super-Resolution

by Zheng Xia, Huizi He, Zixing Zhou, Shanshan Pan and Sai Zhang

Sensors 2026, 26(6), 1992; https://doi.org/10.3390/s26061992 - 23 Mar 2026

Viewed by 552

Abstract

Acoustic imaging, especially ultrasound, underpins a wide range of applications from non-destructive evaluation to medical and materials analysis, yet its performance is ultimately constrained by lateral resolution. This review systematically summarizes recent advances in overcoming diffraction-limited resolution, encompassing traditional focusing techniques, transducer optimization, [...] Read more.

Acoustic imaging, especially ultrasound, underpins a wide range of applications from non-destructive evaluation to medical and materials analysis, yet its performance is ultimately constrained by lateral resolution. This review systematically summarizes recent advances in overcoming diffraction-limited resolution, encompassing traditional focusing techniques, transducer optimization, physical metamaterial lenses, and methods based on algorithmic optimization and deep learning technologies. It comprehensively covers approaches for enhancing acoustic lateral resolution, compares the differences and respective advantages and disadvantages of various methods, and proposes clear directions and recommendations for future research. This work provides robust guidance for subsequent research trends and development opportunities in higher-resolution acoustic imaging. Full article

(This article belongs to the Section Sensing and Imaging)

► Show Figures

Figure 1

13 pages, 2175 KB

Open AccessArticle

Multi-Sensor Measurement of Cylindrical Illuminance

by Michal Kozlok, Marek Balsky and Petr Zak

Sensors 2026, 26(6), 1991; https://doi.org/10.3390/s26061991 - 23 Mar 2026

Viewed by 388

Abstract

Spatial light field metrics, such as cylindrical illuminance, provide essential information for qualitative lighting evaluation, yet they remain far less common in practice than horizontal illuminance. To address this gap, we present a multi-sensor prototype that simultaneously measures horizontal illuminance E_h and [...] Read more.

Spatial light field metrics, such as cylindrical illuminance, provide essential information for qualitative lighting evaluation, yet they remain far less common in practice than horizontal illuminance. To address this gap, we present a multi-sensor prototype that simultaneously measures horizontal illuminance E_h and approximates mean cylindrical illuminance E_z from a set of vertical illuminances uniformly distributed around a cylindrical surface. The device uses a flexible PCB wrapped around a support barrel, along with an inertial and magnetic measurement unit for orientation tracking. The measurements enable direct calculation of the modelling factor defined in the technical standard EN 12 464 and the visualization of the directional light distribution using polar plots and an illuminance solid. Results show that the prototype approximates mean cylindrical illuminance with high accuracy while preserving directional information, allowing the illuminance solid to be decomposed into vector and symmetric components. Compared with conventional approximation methods, the proposed multi-sensor approach reduces spatial error and yields richer data for lighting analysis. These findings indicate that multi-sensor systems can bridge the gap between theoretical spatial metrics and practical photometry and support the improved modelling evaluation and integration of qualitative lighting parameters into routine workflows. Full article

(This article belongs to the Special Issue Methods, Measurements, and Sensor Technology for Lighting Applications)

► Show Figures

Figure 1

26 pages, 5161 KB

Open AccessArticle

LHO-net: A Lightweight Steel Defect Detection Framework Based on Cross-Scale Feature Selection and Adaptive Optimization

by Qi Wang and Haocheng Yan

Sensors 2026, 26(6), 1990; https://doi.org/10.3390/s26061990 - 23 Mar 2026

Viewed by 339

Abstract

To address the issues of poor adaptability to complex scenarios, high computational complexity, and difficulties in terminal deployment of existing steel surface defect detection models, a novel lightweight detection network named LHO-net is proposed, with the Lightweight Multi-Backbone (LM Backbone), the Hierarchical Scale-based [...] Read more.

To address the issues of poor adaptability to complex scenarios, high computational complexity, and difficulties in terminal deployment of existing steel surface defect detection models, a novel lightweight detection network named LHO-net is proposed, with the Lightweight Multi-Backbone (LM Backbone), the Hierarchical Scale-based Pyramid Attention Network (HSPAN), and the Occlusion-aware Detection Head (OAHead). The LM Backbone adopts a dual-branch structure with shared HGStem and a dynamic feature fusion mechanism, effectively capturing multi-dimensional features of irregular defects while extremely compressing model parameters. The HSPAN module realizes efficient fusion of multi-scale features through dynamic feature selection and adaptive upsampling strategies, balancing background noise suppression and defect detail preservation. The OAHead completes adaptive compensation of features in occluded regions by means of deep feature aggregation and exponential normalization technology, significantly enhancing the ability to recognize complex defects. On the NEU-DET dataset, LHO-net achieves a mAP@0.5 of 75.0%, a mAP@0.5:0.95 of 44.0%, and a recall of 73.6%, with a computational complexity of only 2.3 GFLOPS. Compared with the baseline model YOLOv12, it reduces parameters by 64% and computational cost by 60.3%. On the GC-10 dataset, its mAP@0.5 reaches 67.2%, and its detection stability for complex defects such as slender creases and low-contrast water spots is superior to that of mainstream lightweight YOLO variants. Visualization results confirm that the model can effectively avoid common problems such as redundant annotations and false detections and maintains stable recognition performance for various defects. It solves the core contradiction between detection accuracy and lightweight deployment in industrial scenarios, providing an efficient and practical technical solution for real-time steel surface defect detection on resource-constrained terminal devices. Full article

(This article belongs to the Section Fault Diagnosis & Sensors)

► Show Figures

Figure 1

20 pages, 543 KB

Open AccessArticle

EdgeGuard-AI: Zero-Trust and Load-Aware Federated Scheduling for Secure and Low-Latency IoT Edge Networks

by Abdulaziz G. Alanazi and Haifa A. Alanazi

Sensors 2026, 26(6), 1989; https://doi.org/10.3390/s26061989 - 23 Mar 2026

Viewed by 441

Abstract

Edge computing is now widely used to support real-time and safety-critical IoT services. However, current edge schedulers usually optimize only performance, while security verification and trust assessment are handled as separate modules. This separation creates a practical risk: tasks may be assigned to [...] Read more.

Edge computing is now widely used to support real-time and safety-critical IoT services. However, current edge schedulers usually optimize only performance, while security verification and trust assessment are handled as separate modules. This separation creates a practical risk: tasks may be assigned to lightly loaded but compromised edge nodes, or secure nodes may become overloaded, violating latency requirements. We propose EdgeGuard-AI, a unified trust-driven and load-aware scheduling framework inspired by zero-trust security principles for next-generation IoT edge networks. The framework jointly learns dynamic node trust and short-term workload patterns from distributed edge data and integrates both signals into scheduling decisions. Experimental results on a realistic IoT edge security dataset show a task success rate of 97.3 percent, average scheduling latency of 58.1 ms during stress periods, unsafe offloading below 2 percent, and trust discrimination AUC of 0.971. Full article

(This article belongs to the Section Internet of Things)

► Show Figures

Figure 1

14 pages, 3023 KB

Open AccessArticle

Lightweight Stereo Vision for Obstacle Detection and Range Estimation in Micro-Mobility Vehicles

by Jiansheng Ruan, Hui Weng, Zhaojun Yuan, Guangyuan Jin and Liang Zhou

Sensors 2026, 26(6), 1988; https://doi.org/10.3390/s26061988 - 23 Mar 2026

Viewed by 355

Abstract

Micro-mobility vehicles operating in closed, low-speed environments (e.g., parks) require reliable obstacle detection and accurate range estimation under strict constraints on cost, power, and onboard computation. This paper proposes HAGVNet, a lightweight stereo matching network for embedded ranging and validates its practical deployability [...] Read more.

Micro-mobility vehicles operating in closed, low-speed environments (e.g., parks) require reliable obstacle detection and accurate range estimation under strict constraints on cost, power, and onboard computation. This paper proposes HAGVNet, a lightweight stereo matching network for embedded ranging and validates its practical deployability in a target-level ranging pipeline with YOLO11n as the front-end detector. HAGVNet builds a hierarchical attention-guided cost volume (HAGV) that uses coarse-scale geometric priors to modulate fine-scale cost modeling and adopts ConvNeXtV2-style 2D cost aggregation blocks to improve stability and boundary consistency with controlled complexity. For ranging, depth statistics within detected regions are used to estimate target distance and 3D position. The model is pre-trained on SceneFlow and evaluated on KITTI. On SceneFlow, HAGVNet reaches 0.73 px EPE with 20.08 G FLOPs, indicating a favorable accuracy–complexity trade-off under low computation budgets. On an embedded Jetson Orin Nano Super platform, HAGVNet achieves 46.3 FPS under TensorRT FP16, and field tests indicate relative ranging errors of 0.5–8.6% within 2–10 m, demonstrating its practical feasibility for low-speed target-level ranging. Full article

(This article belongs to the Section Sensing and Imaging)

► Show Figures

Figure 1

18 pages, 20418 KB

Open AccessArticle

Localized Query Attack Toward Transformer-Based Visible Object Detectors

by Yang Wang, Ang Li, Zhen Yang and Xunyun Liu

Sensors 2026, 26(6), 1987; https://doi.org/10.3390/s26061987 - 23 Mar 2026

Viewed by 313

Abstract

Transformer-based detectors have demonstrated exceptional accuracy in visible-object detection tasks. However, adversarial patches, specific types of adversarial examples, can disrupt these detectors by introducing unrestricted perturbations into specific image regions. Traditional methodologies focus on placing patches directly on objects and increasing attention scores [...] Read more.

Transformer-based detectors have demonstrated exceptional accuracy in visible-object detection tasks. However, adversarial patches, specific types of adversarial examples, can disrupt these detectors by introducing unrestricted perturbations into specific image regions. Traditional methodologies focus on placing patches directly on objects and increasing attention scores between the patch and all areas of the image to impair detector performance. Nevertheless, these approaches are suboptimal due to significant discrepancies between background and object features, which contradict optimization objectives. Moreover, they overlook the impact of cross-attention mechanisms on detection results. To address these limitations, we introduce a novel approach named Localized Query Attack (LQA), designed to interfere with both self-attention within the encoder and cross-attention in the decoder. Unlike conventional global interference methods, LQA targets object features specifically, enhancing self-attention interactions between the adversarial patch and foreground regions to redirect model focus toward the patch. In the context of decoder cross-attention, we compute the joint attention matrix connecting encoder outputs with object queries. By diminishing the influence of encoder outputs and residual components in this matrix, we amplify the relative importance of the adversarial patch, thereby intensifying the attack’s effectiveness. Our experiments show that LQA achieves an approximately 20% improvement in transfer attack performance compared to the second-best method across various transformer-based detectors. The practical efficacy of LQA is further substantiated through real-world scenario validations, underscoring its applicability. Full article

(This article belongs to the Section Electronic Sensors)

► Show Figures

Figure 1

20 pages, 17836 KB

Open AccessArticle

Temporal Consistency for Reliability Enhancement in Correlation-Based Time–Frequency Domain Reflectometry

by Ju-Bong Lee, Hee Su Lim and Chun-Kwon Lee

Sensors 2026, 26(6), 1986; https://doi.org/10.3390/s26061986 - 22 Mar 2026

Viewed by 389

Abstract

Reflectometry-based sensing systems are widely used in industrial monitoring to assess the condition of distributed assets such as cables and transmission lines. In practical sensing environments, however, correlation-based interpretation can become unreliable because of bilinear interference, dispersive propagation, and excitation mismatch, often producing [...] Read more.

Reflectometry-based sensing systems are widely used in industrial monitoring to assess the condition of distributed assets such as cables and transmission lines. In practical sensing environments, however, correlation-based interpretation can become unreliable because of bilinear interference, dispersive propagation, and excitation mismatch, often producing artifact-related responses that lead to unnecessary inspections and reduced decision reliability. This paper proposes a temporal-consistency-based reliability enhancement framework for correlation-driven time–frequency domain reflectometry (TFDR). Instead of replacing the conventional reflectometry pipeline, the proposed method introduces a reliability-estimation layer that evaluates the trustworthiness of correlation responses and suppresses temporally inconsistent artifacts. Multiple complementary descriptors extracted from the reflected signal are jointly analyzed to determine whether a correlation response is propagation-consistent or more likely to arise from non-physical artifacts. Temporal consistency is modeled using a bidirectional long short-term memory (BiLSTM) architecture that captures long-range dependencies along the propagation sequence. Experimental results obtained from cable reflectometry measurements under varying impedance conditions show that the proposed framework effectively suppresses artifact-related correlation responses while preserving physically meaningful reflections required for fault localization. Additional cross-excitation evaluation provides preliminary evidence that the learned temporal-consistency criterion is not tightly coupled to a single excitation waveform. Because the proposed framework operates as a post-processing reliability layer, it can be integrated into existing reflectometry-based monitoring systems without the modification of the sensing hardware or excitation scheme. Full article

(This article belongs to the Section Fault Diagnosis & Sensors)

► Show Figures

Figure 1

19 pages, 4638 KB

Open AccessArticle

A Training System for Human Standing Stability Using Virtual Viscosity Fields

by Hayato Mikami, Keisuke Shima, Tianyi Wang, Haruto Kai and Koji Shimatani

Sensors 2026, 26(6), 1985; https://doi.org/10.3390/s26061985 - 22 Mar 2026

Viewed by 420

Abstract

Enhancement of postural stability in standing is essential for fall prevention in the context of demographic aging. Against such a background, this study proposes a personalized training system based on individual limits of stability (LOS) for a human standing state. The system evaluates [...] Read more.

Enhancement of postural stability in standing is essential for fall prevention in the context of demographic aging. Against such a background, this study proposes a personalized training system based on individual limits of stability (LOS) for a human standing state. The system evaluates LOS in eight directions using center-of-mass (COM) and center-of-pressure (COP) measurement devices and provides game-based feedback, then promotes balance within the relevant LOS parameters. Loading is individualized by applying greater force to virtual objects as the COP approaches the LOS determined for each subject. Experiments with 32 younger and 19 mature subjects produced evaluations for postural stability index (IPS), LOS area, and COP sway. The results revealed two distinct response patterns: LOS expansion and sway reduction, both observed across younger and mature cohorts. These findings suggest that individualized LOS-based training can be applied to improve standing stability with two distinct strategies. These preliminary findings suggest that individualized LOS-based training is associated with changes in standing stability through two distinct response patterns. Full article

(This article belongs to the Special Issue Sensor Technologies for Mobility in Aging: Development, Validation and Application)

► Show Figures

Figure 1

21 pages, 459 KB

Open AccessArticle

Formation-Constrained Cooperative Localization for UAV Swarms in GNSS-Denied Environments

by Qin Li, Peng Wang, Xiaochun Li, Jieyong Zhang, Ying Luo, Wangsheng Yu and Haiyan Cheng

Sensors 2026, 26(6), 1984; https://doi.org/10.3390/s26061984 - 22 Mar 2026

Viewed by 483

Abstract

Cooperative localization is critical for UAV swarm operations in GNSS-denied environments. The backbone-listener scheme, using a small subset of agents as active backbone nodes and others as passive listeners, offers notable advantages in reducing communication overhead and enhancing swarm scalability. Building on this [...] Read more.

Cooperative localization is critical for UAV swarm operations in GNSS-denied environments. The backbone-listener scheme, using a small subset of agents as active backbone nodes and others as passive listeners, offers notable advantages in reducing communication overhead and enhancing swarm scalability. Building on this scheme, we propose a formation-constrained cooperative localization method to improve accuracy by integrating known formation geometry into the localization process. First, backbone node selection uses a formation-constrained greedy node activation (GNA) strategy with weighted distance fusion, combining measured and ideal formation distances to enable near-optimal selection aligned with formation structure. Second, listener node localization incorporates formation constraints into Chan’s algorithm, paired with angle-of-arrival (AOA) refinement, to ensure estimated positions match expected inter-agent distances. Third, global optimization uses a gradient descent-based refinement to enforce formation constraints across all agent positions. Our theoretical derivations and simulations are limited to the two-dimensional (2D) case. Simulation results validate the proposed method’s improved success rate, reliability, and stability. Its effectiveness is demonstrated across various formation types, with robust adaptability to asymmetric geometries shown to be a valuable feature for practical deployment. Full article

(This article belongs to the Section Navigation and Positioning)

► Show Figures

Figure 1

23 pages, 2900 KB

Open AccessArticle

Laboratory Investigation on the Impact Force of Large Boulders in Debris Flows

by Wei Yi, Bin Yu, Qinghua Liu, Jianchun Hu and Jun Zhou

Sensors 2026, 26(6), 1983; https://doi.org/10.3390/s26061983 - 22 Mar 2026

Viewed by 443

Abstract

The impact of large boulders transported by debris flows is a primary cause of structural damage to mitigation works. However, quantitative modeling remains difficult because of the scarcity of field measurements and the complexity of the flow medium. In this study, a theoretical [...] Read more.

The impact of large boulders transported by debris flows is a primary cause of structural damage to mitigation works. However, quantitative modeling remains difficult because of the scarcity of field measurements and the complexity of the flow medium. In this study, a theoretical model for boulder impact force in debris flows is developed using dimensional analysis based on the Buckingham theorem, subsequently simplified to two dimensionless parameters, and then validated through a series of controlled laboratory experiments. Marble spheres were used as impactors and were released to strike a rigid steel plate under three types of media: clear water, bentonite slurry, and debris flows containing particles of different size classes. The experiments were designed to isolate and quantify the influence of the flow rheology and the suspended solid phase on impact forces. The results show that the impact coefficient c is strongly governed by the debris flow yield stress, bulk density, and the size of suspended particles, following the relationship c = 0.183[τ/(rgd₁)]^−0.1(d/d₀)^0.05. Based on this relationship, a generalized formula for calculating boulder impact forces in debris flows is proposed. The model is further evaluated using field monitoring data from Jiangjiagou, Yunnan Province. The back-calculated boulder diameters fall predominantly within the range of 0.1–0.3 m (76.3–86.8%), which is consistent with field observations. These results indicate that the proposed model captures the essential physical mechanisms governing boulder impacts and provides a rational basis for selecting design parameters in debris flow mitigation engineering. The array-type piezoelectric impact sensing system designed in this study achieves high-precision and high-stability measurement of the impact force of large boulders in debris flows, providing a new sensing technology for debris flow impact monitoring. Full article

(This article belongs to the Topic Advanced Risk Assessment in Geotechnical Engineering)

► Show Figures

Figure 1

22 pages, 3135 KB

Open AccessArticle

Computational Imaging Method for Thermal Infrared Hyperspectral Imaging Based on a Snapshot Divided-Aperture System

by Tianzhen Ma, Zhijing He, Bin Wu, Yutian Lei, Yijie Wang, Xinze Liu, Bingmei Guo, Jiawei Lu, Bo Cheng, Shikai Zan, Chunlai Li and Liyin Yuan

Sensors 2026, 26(6), 1982; https://doi.org/10.3390/s26061982 - 22 Mar 2026

Viewed by 468

Abstract

To address the technical challenge of simultaneously achieving snapshot imaging capability and high spectral resolution in thermal infrared spectral imaging, this paper proposes a computational imaging method based on a snapshot divided-aperture imaging system. In this method, a self-developed divided-aperture snapshot multispectral camera [...] Read more.

To address the technical challenge of simultaneously achieving snapshot imaging capability and high spectral resolution in thermal infrared spectral imaging, this paper proposes a computational imaging method based on a snapshot divided-aperture imaging system. In this method, a self-developed divided-aperture snapshot multispectral camera is utilized to simultaneously capture nine low-spectral-resolution images in a single exposure. The precise registration of the sub-channel images is accomplished via a star-point array calibration method. To construct the spectral reconstruction dataset, a Fourier-transform infrared hyperspectral camera (FTIR HCam) is employed to simultaneously acquire hyperspectral data from real-world scenes. Based on this, a neural network model is applied to reconstruct 127-channel hyperspectral information from the low-dimensional multispectral measurements. Experimental results demonstrate that the proposed method effectively achieves hyperspectral reconstruction while maintaining system compactness and snapshot imaging capability, thus providing a viable technical approach for hyperspectral sensing in dynamic thermal infrared scenarios. Full article

(This article belongs to the Section Sensing and Imaging)

► Show Figures

Figure 1

24 pages, 3552 KB

Open AccessArticle

Optimization of the Spatial Position of the Vibration Acceleration Sensor and the Method of Determining Limit Values in the Diagnostics of Combustion Engine Injection System

by Jan Monieta and Lech Władysław Kasyk

Sensors 2026, 26(6), 1981; https://doi.org/10.3390/s26061981 - 22 Mar 2026

Viewed by 472

Abstract

A new procedure for diagnosing damage to the fuel injection system of marine engines, along with vibration acceleration signal symptoms, is explored with a related built, developed, and tested measuring system. This work fills an important gap given the current lack of a [...] Read more.

A new procedure for diagnosing damage to the fuel injection system of marine engines, along with vibration acceleration signal symptoms, is explored with a related built, developed, and tested measuring system. This work fills an important gap given the current lack of a scientific solution to this problem. A vibration acceleration signal sensor, mounted on a holder elaborated on by the authors, is positioned on the injection pipe between the injection pump and the injector. The output signals from the sensor are sent to an acquisition and analysis system, which is used for processing the signals in the time, amplitude, frequency, and time–frequency domains. Experimental choices, using multiple parameters for a given signal analysis field, are based on the location of the optimal sensor, the direction of the sensor mounting, and the selection of a cumulative diagnostic symptom. The vibration acceleration signals recorded along the injection pipe are found to have the strongest magnitude. This article compares diagnostic values from these signals with previously determined upper and lower limits. As a result, the tested fuel injection system is classified as either able or disabled, using unparalleled tolerance ranges given for both the upper and lower limits. The values of the limits are determined based on the average value for an ability state plus or minus three times the standard error of this mean, which has not been reported in the literature previously. Multiple regression models are developed that relate identified symptoms to the state features of the fuel injection system. In addition, artificial neural networks and machine learning are used to detect developing damage. The probability of correctly classifying the states of the diagnostic parameters is 0.467, alongside a diagnostic accuracy of ≤±4%, with the network correctly classifying the state when the testing accuracy is at least 70.0%. Full article

(This article belongs to the Section Fault Diagnosis & Sensors)

► Show Figures

Graphical abstract

Journal Menu

Journal Browser

Sensors, Volume 26, Issue 6 (March-2 2026) – 271 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI