IF 4 3区综合性期刊

Sensors Pub Date : 2026-10-07 DOI: 10.3390/s26196321

Vadim Morari, Cristian Casut, Daniel Ursu, Eduard V Monaico, Emil V Rusu, Marinela Miclau, Veaceslav V Ursaki

{"title":"Comparative Study of Ti- and Cu-Doped Zn2SnO4 Films Deposited by Spray Pyrolysis for UV Photodetection.","authors":"Vadim Morari, Cristian Casut, Daniel Ursu, Eduard V Monaico, Emil V Rusu, Marinela Miclau, Veaceslav V Ursaki","doi":"10.3390/s26196321","DOIUrl":"10.3390/s26196321","url":null,"abstract":"This study aims to clarify how Ti and Cu doping differently influence the structural, morphological, electrical, and UV photodetection properties of Zn2SnO4 films prepared by spray pyrolysis. The comparison is based on the different chemical and electronic characteristics of the dopants and their resulting effects on charge transport and photoresponse. Zn2SnO4 films doped with 1-3 vol.% Ti or Cu were deposited under identical conditions and characterized using X-ray diffraction, atomic force microscopy, electrical measurements, and transient UV photoresponse at 365 nm. The results demonstrate distinct dopant-dependent structure-property relationships. Ti doping maintains a relatively homogeneous surface morphology and results in higher sheet resistance, whereas Cu doping produces stronger concentration-dependent morphological changes and lower sheet resistance at 1-2 vol.% Cu. The different electrical characteristics are reflected in the UV photodetection performance. The 1 vol.% Ti-doped film exhibits the highest responsivity of 239 mA/W, specific detectivity of 2.73 × 1010 Jones and a recovery time (τ90) of 74 ms, while the 3 vol.% Cu-doped film achieves the highest photo-to-dark current ratio of approximately 763 and a fast recovery time (τ90) of 70 ms. These results demonstrate that Ti and Cu provide different routes for tailoring Zn2SnO4 films for UV photodetection: Ti doping favors high responsivity and detectivity, whereas Cu doping enables efficient dark current suppression and rapid photoresponse. This study therefore establishes a direct relationship between dopant-dependent material properties and UV photodetector performance.","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"26 19","pages":""},"PeriodicalIF":4.0,"publicationDate":"2026-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13667686/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149196378","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multiparametric Regressions and Machine Learning: UAV Imagery for Temperature Prediction of Irrigated Forage Cactus in the Brazilian Semiarid. 多参数回归和机器学习：无人机图像在巴西半干旱地区灌溉牧草仙人掌温度预测中的应用。

IF 4 3区综合性期刊

Sensors Pub Date : 2026-10-07 DOI: 10.3390/s26196327

Eric Gabriel Fernandez Albuquerque da Silva, Thayná Alice Brito Almeida, Raví Emanoel de Melo, João Luis Mendes Pedroso de Lima, José Amilton Santos Júnior, Abelardo Antônio de Assunção Montenegro

{"title":"Multiparametric Regressions and Machine Learning: UAV Imagery for Temperature Prediction of Irrigated Forage Cactus in the Brazilian Semiarid.","authors":"Eric Gabriel Fernandez Albuquerque da Silva, Thayná Alice Brito Almeida, Raví Emanoel de Melo, João Luis Mendes Pedroso de Lima, José Amilton Santos Júnior, Abelardo Antônio de Assunção Montenegro","doi":"10.3390/s26196327","DOIUrl":"10.3390/s26196327","url":null,"abstract":"Vegetation surface temperature is a key indicator of crop water status, particularly in semi-arid regions where water availability constrains agricultural production. This study evaluated whether spectral bands and vegetation indices derived from UAV multispectral imagery can predict the surface temperature of irrigated forage cactus (Opuntia spp.) grown with reused water in the Brazilian semiarid region. A randomized block experiment was conducted in Pesqueira, Pernambuco, Brazil. Canopy temperature was measured with a handheld thermal camera (FLIR E6), and multispectral imagery was acquired with a DJI Phantom 4 UAV across three flights. Linear, polynomial, and multiparametric regression models were compared against four machine learning algorithms (Random Forest (RF), epsilon-support vector regression (SVR), Extreme Gradient Boosting (XGBoost), and K-Nearest Neighbors (KNN)), each evaluated over 1500 randomized train/test splits. Vegetation indices were negatively correlated with canopy temperature, consistent with higher vegetative vigor being associated with cooler canopies. The seven-variable multiparametric regression model achieved the best regression performance (validation R2 = 64.70%, Nash-Sutcliffe efficiency = 0.84), while KNN was the best-performing machine learning algorithm (mean test R2 = 78.11%). The findings suggest that UAV-derived spectral indices represent a promising indirect proxy for estimating forage cactus canopy temperature under the evaluated semiarid conditions. However, independent validation across diverse environments and management conditions is necessary prior to operational implementation as an alternative to direct thermal sensing.","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"26 19","pages":""},"PeriodicalIF":4.0,"publicationDate":"2026-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13667663/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149196401","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multi-Sensor Monitoring of Underground Heritage Sites: A Case Study of the Dragon's Den Cave and St. Kinga's Chapel in Poland. 地下遗址的多传感器监测：以波兰龙穴洞和圣金加教堂为例。

IF 4 3区综合性期刊

Sensors Pub Date : 2026-10-07 DOI: 10.3390/s26196329

Mateusz Jabłoński, Wojciech Jaśkowski

{"title":"Multi-Sensor Monitoring of Underground Heritage Sites: A Case Study of the Dragon's Den Cave and St. Kinga's Chapel in Poland.","authors":"Mateusz Jabłoński, Wojciech Jaśkowski","doi":"10.3390/s26196329","DOIUrl":"10.3390/s26196329","url":null,"abstract":"Long-term structural health monitoring of underground heritage sites is challenging because of high humidity, limited accessibility, complex geological conditions, and aggressive saline environments. This study presents the implementation and long-term assessment of an integrated multi-sensor telemetric monitoring system deployed in two historically significant sites: Dragon's Den Cave beneath Wawel Castle and St. Kinga's Chapel in the Wieliczka Salt Mine. The system combines displacement sensors, strain measurements on rock anchors, and microclimate monitoring to assess structural behavior and support the early detection of potentially hazardous changes. Monitoring was conducted for four years in the cave and two years in the salt mine. Structural measurements were analyzed together with environmental data to distinguish temperature- and humidity-related sensor responses from actual deformation. Periodic geodetic surveys provided independent verification of the telemetric measurements. Both structures remained stable throughout the observation periods. Telemetric measurements recorded changes of up to 0.16 mm at the monitored extensometer locations, while geodetic surveys indicated cumulative displacements of up to approximately 2.6 mm horizontally and 2.4 mm vertically. The results demonstrate that integrating structural and environmental sensing with periodic geodetic verification enables reliable long-term monitoring under demanding underground conditions. The proposed minimally invasive and scalable approach supports preventive conservation and safety management of underground heritage structures.","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"26 19","pages":""},"PeriodicalIF":4.0,"publicationDate":"2026-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13667435/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149196477","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Scooter-Driving Humanoid Robot: Sim-to-Real Transfer Through Deep Reinforcement Learning. 踏板车驾驶人形机器人：通过深度强化学习实现模拟到真实的迁移。

IF 4 3区综合性期刊

Sensors Pub Date : 2026-10-07 DOI: 10.3390/s26196324

Ugo Richard Roux, Saeed Saeedvand, Jacky Baltes

{"title":"Scooter-Driving Humanoid Robot: Sim-to-Real Transfer Through Deep Reinforcement Learning.","authors":"Ugo Richard Roux, Saeed Saeedvand, Jacky Baltes","doi":"10.3390/s26196324","DOIUrl":"10.3390/s26196324","url":null,"abstract":"This paper presents a simulation-to-reality (sim-to-real) transfer process, allowing a full-sized humanoid robot to autonomously balance a two-wheeled scooter and track operator-supplied heading commands through a deep reinforcement learning (DRL) policy, including while carrying a human passenger for the first time. The learned policy governs the coupled balance and steering dynamics only: the heading reference and the target speed are supplied by a human operator, the throttle is set externally, and the initial launch over the first 2 to 3 m and the final stop are performed manually for safety. The system therefore performs autonomous balance and steering-command tracking during motion, rather than autonomous scooter driving. Unlike four-wheeled vehicles where the inherent stability simplifies control, scooter operation demands continuous and precise dynamic balancing coupled with real-time steering control, creating a challenging full-body control task. In this study, we demonstrate a successful integration of DRL techniques to bridge the sim-to-real gap, achieving stable closed-loop control of a humanoid robot balancing and steering a scooter under significant model uncertainty. A multi-scenario real-world evaluation shows the system maintaining control in the demanding low-speed regime of 0.8 to 3.0 m s-1, where balance control is most challenging and where speed varied across the range during trials rather than being held at a set point. Success rates are 0.96 on straight-line driving, 0.92 and 0.72 on light and sharp turns, 0.88 on the traversal of a 60 mm speed bump, and 0.73 while carrying a 70 kg passenger. They fall where sustained precise tracking is required, to 0.28 on a 4.5 m roundabout and 0.33 on part of the standardized Taiwanese scooter license test.","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"26 19","pages":""},"PeriodicalIF":4.0,"publicationDate":"2026-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13667465/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149196617","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

ST-FGDIR: Spatiotemporal Traffic Forecasting with Forecast-Gain-Guided Deviation-Invariance Routing. ST-FGDIR：基于预测增益制导偏差不变路由的时空交通预测。

IF 4 3区综合性期刊

Sensors Pub Date : 2026-10-07 DOI: 10.3390/s26196319

Qianxin Xie, Yancun Jiang, Jinfeng Xu, Yuxuan Zhang

{"title":"ST-FGDIR: Spatiotemporal Traffic Forecasting with Forecast-Gain-Guided Deviation-Invariance Routing.","authors":"Qianxin Xie, Yancun Jiang, Jinfeng Xu, Yuxuan Zhang","doi":"10.3390/s26196319","DOIUrl":"10.3390/s26196319","url":null,"abstract":"Accurate multi-step traffic forecasting requires models to respond to genuine traffic-state transitions while resisting deviations caused by missing or noisy observations. Existing deviation-aware methods characterize departures from recurrent patterns, but deviation magnitude alone does not indicate whether the resulting correction will improve a node-wise forecast. Guided by the insight that correction strength should depend on forecast gain rather than deviation magnitude, we propose Spatiotemporal Traffic Forecasting with Forecast-Gain-Guided Deviation-Invariance Routing (ST-FGDIR). ST-FGDIR constructs an anchor-conditioned stable endpoint and a deviation-corrected candidate through weight-shared current-anchor graph encoding and multi-view deviation evidence. Signed candidate-error differences provide soft forecast-gain targets for a node-wise continuous router, while selective risk extrapolation and a false-activation loss control stable-endpoint risk across perturbation environments and suppress perturbation-induced increases in residual injection. Experiments on METR-LA, PEMS-BAY, PEMS04, and PEMS08 show competitive forecasting performance relative to published baseline results. The inference gates exhibited Spearman correlations of 0.49-0.53 with realized correction gain, and the complete model yielded an average relative MAE increase of 9.38% across 12 controlled dataset-perturbation conditions. These results demonstrate that forecast-gain-guided routing coordinates stable periodic prediction with adaptive deviation correction under both clean and perturbed inputs.","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"26 19","pages":""},"PeriodicalIF":4.0,"publicationDate":"2026-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13667536/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149196866","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Engineering Electrospun PVDF Nanofibers for Next-Generation Wearable Bioelectronics. 工程静电纺PVDF纳米纤维用于下一代可穿戴生物电子学。

IF 4 3区综合性期刊

Sensors Pub Date : 2026-10-07 DOI: 10.3390/s26196328

Reshmi Banerjee, Gymama Slaughter

{"title":"Engineering Electrospun PVDF Nanofibers for Next-Generation Wearable Bioelectronics.","authors":"Reshmi Banerjee, Gymama Slaughter","doi":"10.3390/s26196328","DOIUrl":"10.3390/s26196328","url":null,"abstract":"Electrospun poly(vinylidene fluoride) (PVDF) and P(VDF-TrFE) nanofibers offer a mechanically flexible, high-surface-area platform for wearable bioelectronics, including piezoelectric sensing and self-powered operation. This review establishes a process-structure-property-performance framework for engineering electrospun PVDF- and P(VDF-TrFE)-based wearable systems. Relationships amongst precursor composition, electrospinning parameters, fiber morphology and alignment, crystalline phase formation, dipole orientation, and device-level piezoelectric performance are examined. Electrospinning is compared with other established fabrication methods including solution casting, spin coating, melt processing, phase inversion, and printing in terms of electroactive β-phase fraction, flexibility, scalability, manufacturing complexity, cost, and wearable-device suitability. Device architectures and performance are evaluated across pressure and strain sensing, physiological monitoring, wound-healing interfaces, piezoelectric and triboelectric nanogenerators, and self-powered wearable sensing applications. Additionally, this review considers the distinction between material-level piezoelectric properties and practical system-level performance, alongside challenges in manufacturing reproducibility, realistic mechanical loading, system integration, and long-term operation. Finally, manufacturing and translational considerations are evaluated, and a conceptual engineering roadmap links material selection and processing with microstructure, device architecture, electronics integration, and translation toward robust, scalable wearable bioelectronic systems.","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"26 19","pages":""},"PeriodicalIF":4.0,"publicationDate":"2026-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13667560/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149196554","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SLA-YOLO: A Lightweight SAR Ship Detection Model. SLA-YOLO：一种轻型SAR舰船探测模型。

IF 4 3区综合性期刊

Sensors Pub Date : 2026-10-07 DOI: 10.3390/s26196317

Chaoyue Yin, Nan Bi

{"title":"SLA-YOLO: A Lightweight SAR Ship Detection Model.","authors":"Chaoyue Yin, Nan Bi","doi":"10.3390/s26196317","DOIUrl":"10.3390/s26196317","url":null,"abstract":"Reliable ship detection in synthetic aperture radar (SAR) imagery is challenged by speckles, coastal clutter, scale variation, dense berthing, weak target responses, and limited computing resources. SLA-YOLO is a lightweight YOLO11n detector that coordinates established mechanisms across the backbone, neck, and head. LESCNet combines ECA-enhanced depthwise-pointwise convolution with residual large-kernel refinement; C3k2-DAPE applies input-conditioned branch weighting and channel recalibration at selected fusion blocks; and SA-LSCD shares intermediate transformations while retaining scale-specific predictors. On fixed SSDD and HRSID splits, SLA-YOLO achieves mAP@0.5 values of 97.3% and 90.8% and mAP@0.5:0.95 values of 68.2% and 64.3%, respectively, with 1.70 million parameters and 3.6 GFLOPs. Relative to YOLO11n, parameters and FLOPs decrease by 34.1% and 42.9%, while mAP@0.5 increases by 4.4 and 5.2 percentage points, respectively. A complete 23 ablation shows positive average main effects for all three components and context-dependent interactions. Across four no-fine-tuning transfer settings, SLA-YOLO improves mAP@0.5 by 3.5-6.3 points. Three-seed experiments yield 97.17 ± 0.31% and 90.87 ± 0.35% mAP@0.5 on SSDD and HRSID. Small-ship AP improves over YOLO11n by 10.1 ± 0.5 and 6.9 ± 0.7 points, with larger gains inshore. Network-only throughput reaches 66.8 and 63.4 FPS on an RTX 3090; embedded-device performance remains unverified.","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"26 19","pages":""},"PeriodicalIF":4.0,"publicationDate":"2026-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13667701/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149196827","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Scenario-Based Hardware-in-the-Loop Platform for Real-Time Road-Load Emulation on Production Micromobility Drives. 基于场景的生产微移动驱动实时道路负载仿真硬件在环平台。

IF 4 3区综合性期刊

Sensors Pub Date : 2026-10-07 DOI: 10.3390/s26196320

Assem Meghawer, Martin Võip, Mahmoud Ibrahim, Anton Rassõlkin

{"title":"A Scenario-Based Hardware-in-the-Loop Platform for Real-Time Road-Load Emulation on Production Micromobility Drives.","authors":"Assem Meghawer, Martin Võip, Mahmoud Ibrahim, Anton Rassõlkin","doi":"10.3390/s26196320","DOIUrl":"10.3390/s26196320","url":null,"abstract":"Software-defined electric vehicles have reached micromobility, creating a growing need to evaluate how controller software shapes drive performance and energy use. This requires experimental platforms with quantified emulation fidelity and measurement performance. This paper presents a scenario-based hardware-in-the-loop platform combining a production e-scooter drive, a controlled magnetic particle brake, and a real-time road-load model. The platform reproduces rolling resistance, aerodynamic drag, and speed-gated inertial demand for a nominal 90 kg vehicle. Across eight repetitions of a scaled ECE-15 urban cycle, including a cold first run, the protective load cap retained 99.51% of the commanded load impulse. A post hoc sensitivity analysis of the seven warm runs gave 99.71%. Measured torque and speed tracking root-mean-square errors over all eight runs were 0.704 N·m and 0.593 km/h. Separate electrical characterization produced a machine efficiency map peaking at 86.3% and cycle-energy measurements with a coefficient of variation of 0.72%. Subsampling unfiltered 200 kHz recordings at every sampling offset gave a maximum absolute cycle-energy error of 0.14% at 2 kHz, a hundredfold reduction in retained samples under the tested conditions. These results provide a quantified experimental basis for micromobility drive testing and subsequent controller-software comparisons.","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"26 19","pages":""},"PeriodicalIF":4.0,"publicationDate":"2026-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13667492/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149196381","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Distinguishing Strain Field Visualization Stability from Condition-Level Incremental Strain Accuracy in High-Temperature 2D Digital Image Correlation. 高温二维数字图像相关中应变场可视化稳定性与状态级增量应变精度的区分。

IF 4 3区综合性期刊

Sensors Pub Date : 2026-10-07 DOI: 10.3390/s26196333

Yan Wang, Xiaoshan Ma, Xuan Zhang, Xiaofang Yang, Yu Pei, Huilong Zheng

{"title":"Distinguishing Strain Field Visualization Stability from Condition-Level Incremental Strain Accuracy in High-Temperature 2D Digital Image Correlation.","authors":"Yan Wang, Xiaoshan Ma, Xuan Zhang, Xiaofang Yang, Yu Pei, Huilong Zheng","doi":"10.3390/s26196333","DOIUrl":"10.3390/s26196333","url":null,"abstract":"High-temperature two-dimensional digital image correlation (2D-DIC) can produce spatially unstable strain fields and biased condition-averaged incremental strains, but these error signatures need not respond to the same processing. A smoother strain field is therefore not, by itself, evidence of improved scalar accuracy. We evaluated strain field visualization stability separately from extensometer-referenced, condition-level incremental strain accuracy. An MA956 specimen was tested at 600, 700 and 800 °C under 12 loading conditions spanning 500-2000 µε, with four manually triggered snapshots acquired per condition. The field path used affine registration, pixelwise four-snapshot median fusion and strain radius sensitivity analysis. The scalar path used only the original unfused images and compared four prespecified models under leave-one-temperature-out internal assessment. Increasing the strain radius from 10 to 15 grid points reduced the spatial standard deviation and P95-to-median span by 22.7% and 16.6%, respectively, with little change in scalar RMSE. Controlled simulations with known ground truth showed that local non-affine disturbances primarily increased field dispersion, whereas isotropic scale changes induced approximately common-mode axial and transverse strain drift. Across 20 mechanical strain-magnification conditions, a common-mode-suppressed axial-transverse feature reduced RMSE from 439.7 to 49.8 µε. Guided by this mechanistic evidence, combining the feature with temperature terms yielded the lowest pooled out-of-fold RMSE of 76.27 µε in the experimental data, 69.7% below the raw axial strain baseline. Within the conditions examined, these results demonstrate that improved strain field visualization is insufficient evidence of scalar accuracy and support separate validation of field stability and condition-level incremental strain error.","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"26 19","pages":""},"PeriodicalIF":4.0,"publicationDate":"2026-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13667425/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149196308","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

HCFSL-UAV: Hierarchical Cluster-Head Federated Split Learning for UAV Networks. HCFSL-UAV: UAV网络的分层簇头联合分裂学习。

IF 4 3区综合性期刊

Sensors Pub Date : 2026-10-07 DOI: 10.3390/s26196331

Ruomeng Yuan, Jing Fan, Hao Yu, Yu Peng, Xing Xu

{"title":"HCFSL-UAV: Hierarchical Cluster-Head Federated Split Learning for UAV Networks.","authors":"Ruomeng Yuan, Jing Fan, Hao Yu, Yu Peng, Xing Xu","doi":"10.3390/s26196331","DOIUrl":"10.3390/s26196331","url":null,"abstract":"Resource constraints, time-varying links, and heterogeneous data distributions make distributed model training difficult in unmanned aerial vehicle (UAV) networks. Conventional federated learning can impose substantial computational and communication burdens on resource-constrained UAVs, whereas existing split-learning schemes commonly rely on fixed split points and static participation strategies. To address these limitations, this paper proposes HCFSL-UAV, a hierarchical cluster-head federated split learning framework for UAV networks. The framework adopts a three-layer architecture composed of a central base station, cluster-head UAVs, and member UAVs, and integrates split learning within hierarchical federated aggregation. Member UAVs are dynamically associated with cluster heads according to their own real-time 3-D spatial locations. Within each cluster, participant selection and model splitting are jointly optimized as the set of selected UAVs affects data representativeness and synchronous training latency, while the split point determines the computation, memory, and communication requirements of participating UAVs. A joint optimization problem is formulated to balance data-distribution deviation, training latency, and resource consumption, and an alternating optimization procedure is developed using greedy participant selection and a discrete split-point search. Experiments on FMNIST, CIFAR-10, and CIFAR-100 show that HCFSL-UAV achieves faster convergence and higher test accuracy than the considered federated-learning and split-learning baselines, while reducing communication and computation overhead in the reported settings. These results demonstrate the effectiveness of jointly optimizing participant selection and adaptive model partitioning in dynamic and heterogeneous UAV networks.","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"26 19","pages":""},"PeriodicalIF":4.0,"publicationDate":"2026-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13667644/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149196500","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Sensors最新文献