Sensors

14 pages, 1673 KB

Open AccessArticle

Approximate Analytical Approach for Fast Prediction of Microwave Sensor Response: Numerical Analysis and Results

by Antonio Cuccaro, Raffaele Solimene and Sandra Costanzo

Sensors 2025, 25(18), 5683; https://doi.org/10.3390/s25185683 - 11 Sep 2025

In medical applications, microwave sensors are usually employed to work in direct contact with the human body, therefore requiring an accurate prediction of the electromagnetic interactions with biological tissues. While full-wave simulations can be useful to achieve the above task, they are computationally [...] Read more.

In medical applications, microwave sensors are usually employed to work in direct contact with the human body, therefore requiring an accurate prediction of the electromagnetic interactions with biological tissues. While full-wave simulations can be useful to achieve the above task, they are computationally expensive, especially for iterative sensor optimization. Analytical models may offer a more efficient alternative, but they are often complex, and they must be formulated in a practical way to be useful. As a result, approximate approaches can be advantageous. Traditional approaches, such as plane-wave approximations and transmission-line models, often fail to capture key sensing features. This paper presents an approximate analytical model for standard-aperture sensor configurations to predict the sensor response in terms of the reflection coefficient when placed above a layered medium. The model is based on the assumption that the electromagnetic interaction is primarily governed by the sensor’s dominant mode. Full-wave simulations in the 2–3 GHz frequency range (relevant for medical applications) demonstrate strong agreement with the analytical model, thereby validating its effectiveness as a first-order approximation for sensor–tissue interactions. This provides a reliable and computationally efficient tool to properly manage microwave sensors design in medical applications. Full article

(This article belongs to the Section Electronic Sensors)

► Show Figures

Graphical abstract

25 pages, 5055 KB

Open AccessArticle

A Hybrid Model Based on a Dual-Attention Mechanism for the Prediction of Remaining Useful Life of Aircraft Engines

by Chenwen He, Zixiang Li, Chenyu Zheng, Zikai Zhang and Liping Zhang

Sensors 2025, 25(18), 5682; https://doi.org/10.3390/s25185682 - 11 Sep 2025

Abstract

Estimating the Remaining Useful Life (RUL) of aircraft engines plays a vital role in the field of prognostics and health management. In multi-dimensional time series regression tasks, accurately capturing both time series features and sensor features, as well as integrating these two types [...] Read more.

Estimating the Remaining Useful Life (RUL) of aircraft engines plays a vital role in the field of prognostics and health management. In multi-dimensional time series regression tasks, accurately capturing both time series features and sensor features, as well as integrating these two types of features, poses a significant challenge for RUL prediction. The sensor features represent the weights of each sensor on the RUL prediction results. To overcome this challenge, we introduce a hybrid model based on a dual-attention mechanism. Initially, a temporal feature extraction block is applied to map the time-step dimension into a hidden representation space, facilitating the capture of complex temporal dynamics. These patterns are then refined using a multi-head self-attention mechanism. Subsequently, a sensor feature extraction block is applied to capture sensor-specific characteristics. Each sensor sequence is treated as a separate channel, compressed to derive sensor weights, and integrated to form global features that fuse temporal and sensor-level representations. Finally, RUL is estimated via a regression layer. The proposed method is demonstrated to be effective on the Commercial Modular Aero-Propulsion System Simulation (C-MAPSS) dataset. Compared with the state-of-the-art CTNet model, the proposed method achieves 7% and 9% gains in RMSE and Score, respectively, on the FD001 dataset. Full article

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

► Show Figures

Figure 1

25 pages, 4235 KB

Open AccessArticle

A Performance Study of Deep Neural Network Representations of Interpretable ML on Edge Devices with AI Accelerators

by Julian Schauer, Payman Goodarzi, Jannis Morsch and Andreas Schütze

Sensors 2025, 25(18), 5681; https://doi.org/10.3390/s25185681 - 11 Sep 2025

Abstract

With the rising adoption of machine learning (ML) and deep learning (DL) applications, the demand for deploying these algorithms closer to sensors has grown significantly, particularly in sensor-driven use cases such as predictive maintenance (PM) and condition monitoring (CM). This study investigated a [...] Read more.

With the rising adoption of machine learning (ML) and deep learning (DL) applications, the demand for deploying these algorithms closer to sensors has grown significantly, particularly in sensor-driven use cases such as predictive maintenance (PM) and condition monitoring (CM). This study investigated a novel application-oriented approach to representing interpretable ML inference as deep neural networks (DNNs) regarding the latency and energy efficiency on the edge, to tackle the problem of inefficient, high-effort, and uninterpretable-implementation ML algorithms. For this purpose, the interpretable deep neural network representation (IDNNRep) was integrated into an open-source interpretable ML toolbox to demonstrate the inference time and energy efficiency improvements. The goal of this work was to enable the utilization of generic artificial intelligence (AI) accelerators for interpretable ML algorithms to achieve efficient inference on edge hardware in smart sensor applications. This novel approach was applied to one regression and one classification task from the field of PM and validated by implementing the inference on the neural processing unit (NPU) of the QXSP-ML81 Single-Board Computer and the tensor processing unit (TPU) of the Google Coral. Different quantization levels of the implementation were tested against common Python and C++ implementations. The novel implementation reduced the inference time by up to 80% and the mean energy consumption by up to 76% at the lowest precision with only a 0.4% loss of accuracy compared to the C++ implementation. With the successful utilization of generic AI accelerators, the performance was further improved with a 94% reduction for both the inference time and the mean energy consumption. Full article

(This article belongs to the Section Intelligent Sensors)

► Show Figures

Figure 1

20 pages, 6006 KB

Open AccessArticle

A GRU-KAN Surrogate Model with Genetic Algorithm Uniform Sampling for Active Magnetic Bearings–Rotor Critical Speed Prediction

by Jiahang Cui, Jianghong Li, Feichao Cai, Zhenmin Zhao and Yuxi Liu

Sensors 2025, 25(18), 5680; https://doi.org/10.3390/s25185680 - 11 Sep 2025

Abstract

With the development of active magnetic bearings (AMBs) toward higher speeds, understanding high-speed rotor dynamics has become a crucial focus in AMB research. Traditional finite element modeling (FEM) methods, however, are unable to rapidly and comprehensively uncover the complex interplay between controller parameters [...] Read more.

With the development of active magnetic bearings (AMBs) toward higher speeds, understanding high-speed rotor dynamics has become a crucial focus in AMB research. Traditional finite element modeling (FEM) methods, however, are unable to rapidly and comprehensively uncover the complex interplay between controller parameters and dynamic behavior. To address this limitation, a surrogate modeling approach based on a hybrid gated recurrent unit–Kolmogorov–Arnold network (GRU-KAN) is introduced to mathematically capture the effects of coupled control gains on rotor dynamics. To enhance model generalization, a genetic algorithm-driven uniform design sampling strategy is also implemented. Comparative studies against support vector regression and Kriging surrogates indicate a higher coefficient of determination (

R^{2} = 0.9887

) and lower residuals for the proposed approach. Experimental validation across multiple controller parameter combinations shows that the resulting machine learning surrogate predicts the critical speed with a mean absolute error of only 38.51 rpm and a mean absolute percentage error of

1.56 \times 10^{- 1} %

, while requiring merely

1.14 \times 10^{- 4}

s per evaluation—compared to 201 s for traditional FEM. These findings demonstrate the surrogate’s efficiency, accuracy, and comprehensive predictive capabilities, offering an effective method for rapid critical speed estimation in AMB–rotor systems. Full article

(This article belongs to the Section Physical Sensors)

44 pages, 7171 KB

Open AccessArticle

UniROS: ROS-Based Reinforcement Learning Across Simulated and Real-World Robotics

by Jayasekara Kapukotuwa, Brian Lee, Declan Devine and Yuansong Qiao

Sensors 2025, 25(18), 5679; https://doi.org/10.3390/s25185679 - 11 Sep 2025

Abstract

Reinforcement Learning (RL) enables robots to learn and improve from data without being explicitly programmed. It is well-suited for tackling complex and diverse robotic tasks, offering adaptive solutions without relying on traditional, hand-designed approaches. However, RL solutions in robotics have often been confined [...] Read more.

Reinforcement Learning (RL) enables robots to learn and improve from data without being explicitly programmed. It is well-suited for tackling complex and diverse robotic tasks, offering adaptive solutions without relying on traditional, hand-designed approaches. However, RL solutions in robotics have often been confined to simulations, with challenges in transferring the learned knowledge or learning directly in the real world due to latency issues, lack of a standardized structure, and complexity of integration with real robot platforms. While the use of Robot Operating System (ROS) provides an advantage in addressing these challenges, existing ROS-based RL frameworks typically support sequential, turn-based agent-environment interactions, which fail to represent the continuous, dynamic nature of real-time robotics or support robust multi-robot integration. This paper addresses this gap by proposing UniROS, a novel ROS-based RL framework explicitly designed for real-time multi-robot/task applications. UniROS introduces a ROS-centric implementation strategy for creating RL environments that support asynchronous and concurrent processing, which is pivotal in reducing the latency between agent-environment interactions. This study validates UniROS through practical robotic scenarios, including direct real-world learning, sim-to-real policy transfer, and concurrent multi-robot/task learning. The proposed framework, including all examples and supporting packages developed in this study, is publicly available on GitHub, inviting wider use and exploration in the field. Full article

(This article belongs to the Section Sensors and Robotics)

► Show Figures

Figure 1

13 pages, 531 KB

Open AccessArticle

Automated Remote Detection of Falls Using Direct Reconstruction of Optical Flow Principal Motion Parameters

by Simeon Karpuzov, Stiliyan Kalitzin, Olga Georgieva, Alex Trifonov, Tervel Stoyanov and George Petkov

Sensors 2025, 25(18), 5678; https://doi.org/10.3390/s25185678 - 11 Sep 2025

Abstract

Detecting and alerting for falls is a crucial component of both healthcare and assistive technologies. Wearable devices are vulnerable to damage and require regular inspection and maintenance. Manned video surveillance avoids these problems, but it involves constant labor-intensive attention and, in most cases, [...] Read more.

Detecting and alerting for falls is a crucial component of both healthcare and assistive technologies. Wearable devices are vulnerable to damage and require regular inspection and maintenance. Manned video surveillance avoids these problems, but it involves constant labor-intensive attention and, in most cases, may interfere with the privacy of the observed individuals. To address this issue, in this work we introduce and evaluate a novel approach for fully automated fall detection. The presented technique uses direct reconstruction of principal motion parameters, avoiding the computationally expensive full optical flow reconstruction and still providing relevant descriptors for accurate detections. Our method is systematically compared with state-of-the-art techniques. Comparisons of detection accuracy, computational efficiency, and suitability for real-time applications are presented. Experimental results demonstrate notable improvements in accuracy while maintaining a lower computational cost compared to traditional methods, making our approach highly adaptable for real-world deployment. The findings highlight the robustness and universality of our model, suggesting its potential for integration into broader surveillance technologies. Future directions for development will include optimization for resource-constrained environments and deep learning enhancements to refine detection precision. Full article

(This article belongs to the Special Issue Advanced Signal and Image Processing Techniques for Sensor Applications)

19 pages, 20851 KB

Open AccessArticle

A Wavelet-Recalibrated Semi-Supervised Network for Infrared Small Target Detection Under Data Scarcity

by Cheng Jiang, Jingwen Ma, Xinpeng Zhang, Chiming Tong, Zhongqi Ma and Yongshi Jie

Sensors 2025, 25(18), 5677; https://doi.org/10.3390/s25185677 - 11 Sep 2025

Abstract

Infrared small target detection has long faced significant challenges due to the extremely small size of targets, low contrast, and the scarcity of annotated data. To tackle these issues, we propose a wavelet-recalibrated semi-supervised network (WRSSNet) that integrates synthetic data augmentation, feature reconstruction, [...] Read more.

Infrared small target detection has long faced significant challenges due to the extremely small size of targets, low contrast, and the scarcity of annotated data. To tackle these issues, we propose a wavelet-recalibrated semi-supervised network (WRSSNet) that integrates synthetic data augmentation, feature reconstruction, and semi-supervised learning, aiming to fully exploit the potential of unlabeled infrared images under limited supervision. We construct a dataset containing 843 visible-light small target images and employ an improved CycleGAN model to convert them into high-quality pseudo-infrared images, effectively expanding the scale of training data for infrared small target detection. In addition, we design a lightweight wavelet-enhanced channel recalibration and fusion (WECRF) module, which integrates wavelet decomposition with both channel and spatial attention mechanisms. This module enables adaptive reweighting and efficient fusion of multi-scale features, highlighting high-frequency details and weak target responses. Extensive experiments on two public infrared small target datasets, NUAA-SIRST and IRSTD-1K, demonstrate that WRSSNet achieves superior detection accuracy and lower false alarm rates compared to several state-of-the-art methods, while maintaining low computational complexity. Full article

(This article belongs to the Special Issue Target Detection, Tracking and Identification Using Multi-Sensor Systems)

17 pages, 3815 KB

Open AccessArticle

LMeRAN: Label Masking-Enhanced Residual Attention Network for Multi-Label Chest X-Ray Disease Aided Diagnosis

by Hongping Fu, Chao Song, Xiaolong Qu, Dongmei Li and Lei Zhang

Sensors 2025, 25(18), 5676; https://doi.org/10.3390/s25185676 - 11 Sep 2025

Abstract

Chest X-ray (CXR) imaging is essential for diagnosing thoracic diseases, and computer-aided diagnosis (CAD) systems have made substantial progress in automating the interpretation of CXR images. However, some existing methods often overemphasize local features while neglecting global context, limiting their ability to capture [...] Read more.

Chest X-ray (CXR) imaging is essential for diagnosing thoracic diseases, and computer-aided diagnosis (CAD) systems have made substantial progress in automating the interpretation of CXR images. However, some existing methods often overemphasize local features while neglecting global context, limiting their ability to capture the broader pathological landscape. Moreover, most methods fail to model label correlations, leading to insufficient utilization of prior knowledge. To address these limitations, we propose a novel multi-label CXR image classification framework, termed the Label Masking-enhanced Residual Attention Network (LMeRAN). Specifically, LMeRAN introduces an original label-specific residual attention to capture disease-relevant information effectively. By integrating multi-head self-attention with average pooling, the model dynamically assigns higher weights to critical lesion areas while retaining global contextual features. In addition, LMeRAN employs a label mask training strategy, enabling the model to learn complex label dependencies from partially available label information. Experiments conducted on the large-scale public dataset ChestX-ray14 demonstrate that LMeRAN achieves the highest mean AUC value of 0.825, resulting in an increase of 3.1% to 8.0% over several advanced baselines. To enhance interpretability, we also visualize the lesion regions relied upon by the model for classification, providing clearer insights into the model’s decision-making process. Full article

(This article belongs to the Special Issue Vision- and Image-Based Biomedical Diagnostics—2nd Edition)

► Show Figures

Figure 1

22 pages, 20760 KB

Open AccessArticle

Multi-Camera 3D Digital Image Correlation with Pointwise-Optimized Model-Based Stereo Pairing

by Wenxiang Qin, Feiyue Wang, Shaopeng Hu, Kohei Shimasaki and Idaku Ishii

Sensors 2025, 25(18), 5675; https://doi.org/10.3390/s25185675 - 11 Sep 2025

Abstract

Dynamic deformation measurement (DDM) is critical across infrastructure and industrial applications. Among various advanced techniques, multi-camera digital image correlation (MC-DIC) stands out due to its ability to achieve wide-range, full-field, and non-contact 3D DDM by pairing camera subsystems. However, existing MC-DIC methods typically [...] Read more.

Dynamic deformation measurement (DDM) is critical across infrastructure and industrial applications. Among various advanced techniques, multi-camera digital image correlation (MC-DIC) stands out due to its ability to achieve wide-range, full-field, and non-contact 3D DDM by pairing camera subsystems. However, existing MC-DIC methods typically rely on inefficient manual pairing or a simplistic strategy that aggregates all visible cameras for measuring specific object regions, leading to camera over-grouping. These limitations often result in cumbersome system setup and ill-measured deformations. To overcome these challenges, we propose a novel MC-DIC method with pointwise-optimized model-based stereo pairing (MPMC-DIC). By automatically evaluating and selecting camera pairs based on five evaluation factors derived from 3D model and calibrated cameras, the proposed method overcomes the over-grouping problem and achieves high-precision DDM of semi-rigid objects. A 5 × 5 cm cylinder experiment demonstrated an accuracy of 0.03 mm for both horizontal and depth displacements in the 0.0–5.0 mm range, and validated strong robustness against cluttered backgrounds using a 2 × 4 camera array. Vibration measurement of a 9 × 15 × 16 cm PC speaker operating at 50 Hz, using eight surrounding cameras capturing 1920 × 1080 images at 400 fps, confirmed the proposed method’s capability to perform wide-range dynamic deformation analysis and its robustness against complex object geometries. Full article

(This article belongs to the Special Issue Industrial Soft-Sensing Technology Based on Data-Driven and Artificial Intelligence Technologies)

24 pages, 2701 KB

Open AccessArticle

A Scheduling Method for Maintenance Tasks of Damaged Equipment Based on Digital Twin and Robust Optimization

by Mingjie Jiang, Tiejun Jiang, Lijun Guo and Shaohua Liu

Sensors 2025, 25(18), 5674; https://doi.org/10.3390/s25185674 - 11 Sep 2025

Abstract

Aiming at the problems that traditional maintenance task scheduling schemes for damaged equipment have, poor adaptability to changes in uncertain factors and difficult-to-deal-with emergency scenarios, this paper proposes a maintenance task scheduling method for battle-damaged equipment based on digital twin (DT) and robust [...] Read more.

Aiming at the problems that traditional maintenance task scheduling schemes for damaged equipment have, poor adaptability to changes in uncertain factors and difficult-to-deal-with emergency scenarios, this paper proposes a maintenance task scheduling method for battle-damaged equipment based on digital twin (DT) and robust optimization. The purpose is to realize the dynamic synchronization between physical entities and virtual models through DT technology, and to leverage the anti-interference characteristics of robust optimization. The method involves constructing a multi-objective optimization model that maximizes the comprehensive importance of damaged equipment and minimizes maintenance time, and solving the model using the discrete particle swarm optimization (DPSO) algorithm. Simulation results show that this method can improve the efficiency of maintenance scheduling and the anti-interference ability in emergency situations. Through the comparison of three indicators, DT-DPSO performs the best in the maintenance scheduling of battle-damaged equipment: its convergence speed is 33.3% faster than that of DPSO and 20% faster than that of DT-non-dominated sorting genetic algorithm II (DT-NSGAII); its robustness is 16.3% higher than that of DPSO and 10.7% higher than that of DT-NSGAII; its dynamic reallocation speed is more than 40% faster than that of DPSO and more than 30% faster than that of DT-NSGAII. This method is suitable for maintenance scheduling requirements of high speed, stability, and anti-interference. Full article

(This article belongs to the Section Industrial Sensors)

► Show Figures

Figure 1

18 pages, 1645 KB

Open AccessArticle

Validation of Inertial Measurement Units for Measuring Lower-Extremity Kinematics During Squat–Pivot and Stoop–Twist Lifting Tasks

by Rutuja A. Kulkarni, Rajit Banerjee, Vicki Z. Wang, Marcel Oliart, Verity Rampulla, Prithvi Das and Alicia M. Koontz

Sensors 2025, 25(18), 5673; https://doi.org/10.3390/s25185673 - 11 Sep 2025

Abstract

Optokinetic motion capture (OMC) is the gold standard for measuring the kinematics associated with lifting posture. Unfortunately, limitations exist, including cost, portability, and marker occlusion. The purpose of this study is to evaluate the agreement between OMC and inertial measurement units (IMUs) for [...] Read more.

Optokinetic motion capture (OMC) is the gold standard for measuring the kinematics associated with lifting posture. Unfortunately, limitations exist, including cost, portability, and marker occlusion. The purpose of this study is to evaluate the agreement between OMC and inertial measurement units (IMUs) for quantifying joint kinematics during squat–pivot and stoop–twist lifting tasks. Ten unimpaired adults wearing both IMUs and OMC markers performed 24 lifting trials. Correlation coefficients and Root Mean Square Error (RMSE) between IMU and OMC time-series signals were computed for trunk and lower-extremity joints. Peak values obtained from each system during each trial were analyzed via Bland–Altman plots. Results show high correlations for trunk, knee, and ankle flexion angles (>0.9) and ankle rotation angles (>0.7). Moderate correlation was found for trunk axial rotation and lateral flexion angles (0.5–0.7). RMSE was under 9° for each angle. Biases between systems ranged from 0.3° to 16°. Both systems were able to detect statistically significant differences in peak angles between the two postures (p < 0.05). IMUs show promise for recording field data on complex lifting tasks. Full article

(This article belongs to the Special Issue Wearable Inertial Sensors for Human Movement Analysis)

► Show Figures

Figure 1

24 pages, 3058 KB

Open AccessArticle

Research on LiDAR-Assisted Optimization Algorithm for Terrain-Aided Navigation of eVTOL.

by Guangming Zhang, Jing Zhou, Zhonghang Duan and Weiwei Zhao

Sensors 2025, 25(18), 5672; https://doi.org/10.3390/s25185672 - 11 Sep 2025

Abstract

To address the high-precision navigation requirements of urban low-altitude electric vertical take-off and landing (eVTOL) aircraft in environments where global navigation satellite systems (GNSSs) are denied and under complex urban terrain conditions, a terrain-matching optimization algorithm based on light detection and ranging (LiDAR) [...] Read more.

To address the high-precision navigation requirements of urban low-altitude electric vertical take-off and landing (eVTOL) aircraft in environments where global navigation satellite systems (GNSSs) are denied and under complex urban terrain conditions, a terrain-matching optimization algorithm based on light detection and ranging (LiDAR) is proposed. Given the issues of GNSS signal susceptibility to occlusion and interference in urban low-altitude environments, as well as the error accumulation in inertial navigation systems (INSs), this algorithm leverages LiDAR point cloud data to assist in constructing a digital elevation model (DEM). A terrain-matching optimization algorithm is then designed, incorporating enhanced feature description for key regions and an adaptive random sample consensus (RANSAC)-based misalignment detection mechanism. This approach enables efficient and robust terrain feature matching and dynamic correction of INS positioning errors. The simulation results demonstrate that the proposed algorithm achieves a positioning accuracy better than 2 m in complex scenarios such as typical urban canyons, representing a significant improvement of 25.0% and 31.4% compared to the traditional SIFT-RANSAC and SURF-RANSAC methods, respectively. It also elevates the feature matching accuracy rate to 90.4%; meanwhile, at a 95% confidence level, the proposed method significantly increases the localization success rate to 96.8%, substantially enhancing the navigation and localization accuracy and robustness of eVTOLs in complex low-altitude environments. Full article

(This article belongs to the Section Navigation and Positioning)

21 pages, 5659 KB

Open AccessArticle

PMMCT: A Parallel Multimodal CNN-Transformer Model to Detect Slow Eye Movement for Recognizing Driver Sleepiness

by Yingying Jiao, Jiajia Zhang and Zhuqing Jiao

Sensors 2025, 25(18), 5671; https://doi.org/10.3390/s25185671 - 11 Sep 2025

Abstract

Sleepiness at the wheel is an important contributor to road traffic accidents. Slow eye movement (SEM) serves as a reliable physiological indicator for the sleep onset period (SOP). To detect SEM for recognizing drivers’ SOP, a Parallel Multimodal CNN-Transformer (PMMCT) model is proposed. [...] Read more.

Sleepiness at the wheel is an important contributor to road traffic accidents. Slow eye movement (SEM) serves as a reliable physiological indicator for the sleep onset period (SOP). To detect SEM for recognizing drivers’ SOP, a Parallel Multimodal CNN-Transformer (PMMCT) model is proposed. The model employs two parallel feature extraction modules to process bimodal signals, each comprising convolutional layers and Transformer encoder layers. The extracted features are fused and then classified using fully connected layers. The model is evaluated on two bimodal signal combinations HEOG + O2 and HEOG + HSUM, where HSUM is the sum of two single-channel horizontal electrooculogram (HEOG) signals and captures electroencephalograph (EEG) features similar to those in the conventional O2 channel. Experimental results indicate that using the PMMCT model, the HEOG + HSUM combination performs comparably to the HEOG + O2 combination and outperforms unimodal HEOG by 2.73% in F1-score, with average classification accuracy and F1-score of 99.89% and 99.35%, outperforming CNN, CNN-LSTM, and CNN-LSTM-Attention models. The model exhibits minimal false positives and false negatives, with average values of 5.2 and 0.8. By combining CNNs’ local feature extraction with Transformers’ global temporal modeling, and using only two HEOG electrodes, the system offers superior performance while enhancing wearable device comfort for real-world applications. Full article

(This article belongs to the Section Biomedical Sensors)

16 pages, 2826 KB

Open AccessArticle

Research on Multi-Sensor Fusion Localization for Forklift AGV Based on Adaptive Weight Extended Kalman Filter

by Qiang Wang, Junqi Wu, Yinghua Liao, Bo Huang, Hang Li and Jiajun Zhou

Sensors 2025, 25(18), 5670; https://doi.org/10.3390/s25185670 - 11 Sep 2025

Abstract

This study addresses the problem localization deviation caused by cumulative wheel odometry errors in Automated Guided Vehicles (AGVs) operating in complex environments by proposing an adaptive localization method based on multi-sensor fusion. Within an Extended Kalman Filter (EKF) framework, the proposed approach integrates [...] Read more.

This study addresses the problem localization deviation caused by cumulative wheel odometry errors in Automated Guided Vehicles (AGVs) operating in complex environments by proposing an adaptive localization method based on multi-sensor fusion. Within an Extended Kalman Filter (EKF) framework, the proposed approach integrates internal sensor predictions with external positioning data corrections, employing an adaptive weighting algorithm to dynamically adjust the contributions of different sensors. This effectively suppresses errors induced by factors such as ground friction and uneven terrain. The experimental results demonstrate that the method achieves a localization accuracy of 13 mm, and the simulation results show a higher accuracy of 10 mm under idealized conditions. The minor discrepancy is attributed to unmodeled noise and systematic errors in the complex real-world environment, thus validating the robustness of the proposed approach while maintaining robustness against challenges such as Non-Line-of-Sight (NLOS) obstructions and low-light conditions. The synergistic combination of LiDAR and odometry not only ensures data accuracy but also enhances system stability, providing a reliable navigation solution for AGVs in industrial settings. Full article

(This article belongs to the Section Navigation and Positioning)

► Show Figures

Figure 1

21 pages, 1557 KB

Open AccessArticle

Spectral-Based Fault Detection Method in Marine Diesel Engine Operation

by Joško Radić, Matko Šarić and Ante Rubić

Sensors 2025, 25(18), 5669; https://doi.org/10.3390/s25185669 - 11 Sep 2025

Abstract

The possibility of developing autonomous vessels has recently become increasingly interesting. As most vessels are powered by diesel engines, the idea of developing a method to detect engine malfunctions by analyzing signals from microphones placed near the engine and accelerometers mounted on the [...] Read more.

The possibility of developing autonomous vessels has recently become increasingly interesting. As most vessels are powered by diesel engines, the idea of developing a method to detect engine malfunctions by analyzing signals from microphones placed near the engine and accelerometers mounted on the engine housing is intriguing. This paper presents a method for detecting engine malfunctions by analyzing signals obtained from the output of a microphone and accelerometer. The algorithm is based on signal analysis in the frequency domain using discrete Fourier transform (DFT), and the same procedure is applied to both acoustic and vibration data. The proposed method was tested on a six-cylinder marine diesel engine where a fault was emulated by deactivating one cylinder. In controlled experiments across five rotational speeds, the method achieved an accuracy of approximately 98.3% when trained on 75 operating cycles and evaluated over 15 cycles. The average precision and recall across all sensors exceeded 97% and 96%, respectively. The ability of the algorithm to treat microphone and accelerometer signals identically simplifies implementation, and the detection accuracy can be increased further by adding additional sensors. Full article

(This article belongs to the Special Issue Vibration Engineering, Reliability Assessment and Fault Diagnosis in Mechanical Systems)

► Show Figures

Figure 1

25 pages, 1484 KB

Open AccessArticle

A Hybrid Approach for IoT Security: Combining Ensemble Learning with Fuzzy Logic

by Aykut Karakaya

Sensors 2025, 25(18), 5668; https://doi.org/10.3390/s25185668 - 11 Sep 2025

Abstract

The rapid expansion of Internet of Things (IoT) devices has led to substantial progress in various fields. The diverse and resource-limited characteristics of IoT devices make them susceptible to numerous cyber threats, especially malware. Traditional security approaches fall short of effectively addressing these [...] Read more.

The rapid expansion of Internet of Things (IoT) devices has led to substantial progress in various fields. The diverse and resource-limited characteristics of IoT devices make them susceptible to numerous cyber threats, especially malware. Traditional security approaches fall short of effectively addressing these challenges. In this paper, a novel hybrid approach based on the integration of ensemble learning and fuzzy logic is proposed to enhance IoT security. While the ensemble learning model combines multiple classifiers to improve detection accuracy, fuzzy logic enables a more flexible and interpretable assessment of the security status of IoT systems. Experimental results reveal that the proposed framework provides high-accuracy malware detection and, through the fuzzy system built upon the rule base derived from the ensemble model, offers a more flexible and human intuition-oriented evaluation capability. This study offers an effective solution for ensuring IoT system security, providing an applicable approach across diverse IoT ecosystems. Full article

(This article belongs to the Section Internet of Things)

► Show Figures

Figure 1

30 pages, 14058 KB

Open AccessArticle

Effect of Imaging Range on Performance of Terahertz Coded-Aperture Imaging

by Yan Teng, Haodong Yang, Xinhong Cui, Xiaoze Li and Yanchao Shi

Sensors 2025, 25(18), 5667; https://doi.org/10.3390/s25185667 - 11 Sep 2025

Abstract

This paper reveals a counterintuitive, non-monotonic dependence of terahertz coded-aperture imaging (TCAI) performance on the imaging range. This phenomenon stems from phase-induced spatiotemporal correlations in the reference-signal matrix (RSM), governed by the wavefront phase interactions between the coded-aperture elements and scatterers on the [...] Read more.

This paper reveals a counterintuitive, non-monotonic dependence of terahertz coded-aperture imaging (TCAI) performance on the imaging range. This phenomenon stems from phase-induced spatiotemporal correlations in the reference-signal matrix (RSM), governed by the wavefront phase interactions between the coded-aperture elements and scatterers on the imaging plane. Image quality deteriorates noticeably when a specific dimensionless criterion, which is defined mathematically and physically in this work, precisely reaches integer values. Under such conditions, the relative phase difference concentrates or clusters into discrete values determined by the imaging range, leading to strong column and row correlations in RSM that compromise the spatiotemporal independence essential for high-quality reconstruction. For imaging ranges exceeding the critical threshold determined by the number of grid points along one dimension of the imaging plane, two degradation mechanisms emerge: increased correlation between RSM columns mapping to directly adjacent scatterers and phase coverage reduction in wavefront encoding. Both effects intensify as the imaging range increases, resulting in a monotonic deterioration of imaging performance. Crucially, reconstruction fails primarily when strong correlations involve dominant scatterers, whereas correlations among non-dominant (dummy) scatterers have a negligible impact. The Two-step Iterative Shrinkage/Thresholding (TwIST) algorithm demonstrates superior robustness under these challenging conditions compared to some other conventional methods. These insights provide practical guidance for optimizing TCAI system design and operational range selection to avoid performance degradation zones. Full article

(This article belongs to the Section Sensing and Imaging)

► Show Figures

Figure 1

18 pages, 3170 KB

Open AccessArticle

Revealing Lunar Far-Side Polarization Characteristics via FeO Abundance Distribution Correlations with Ground-Based Polarimetric Data

by Hanlin Ye, Weinan Wang, Jinsong Ping and Yin Jin

Sensors 2025, 25(18), 5666; https://doi.org/10.3390/s25185666 - 11 Sep 2025

Abstract

Due to the tidal locking, the far side of the Moon is permanently turned away from the Earth. Its polarization characteristics are still poorly understood, limiting our knowledge of material composition and evolution. Previous studies have indicated a correlation between the distributions of [...] Read more.

Due to the tidal locking, the far side of the Moon is permanently turned away from the Earth. Its polarization characteristics are still poorly understood, limiting our knowledge of material composition and evolution. Previous studies have indicated a correlation between the distributions of degree of polarization (DOP) and the iron oxide (FeO) abundance on the Moon, suggesting a new approach to infer the polarization characteristics of the lunar far side from FeO abundance distribution. Three critical issues have been analyzed: (1) A linear regression model between DOP and FeO abundance is proposed based on control points from ground-based near side polarization images. (2) The DOP distribution of the lunar far side is estimated, based on the established model, revealing significant hemispheric differences in polarization characteristics. (3) The relationship between DOP and lunar phase angle is examined, with the fitted values demonstrating strong agreement with the observations in both magnitude and variation trend. These insights offer valuable guidance for comprehensive polarimetric studies of the Moon. Full article

(This article belongs to the Special Issue Advanced Subsystems and Technologies for Space and Airborne Communication Networks)

► Show Figures

Figure 1

17 pages, 2754 KB

Open AccessArticle

Exploring the Determinants and Correlates of Health-Enhancing Physical Activity of Adults in Eastern Poland

by Marian J. Stelmach, Joanna Baj-Korpak, Ewelina Niźnikowska, Barbara Bergier, Michał Bergier, Dorota Tomczyszyn, Adam Szepeluk and Paulo Rocha

Sensors 2025, 25(18), 5665; https://doi.org/10.3390/s25185665 - 11 Sep 2025

Abstract

In Poland—especially in the less developed eastern regions—the level of health-enhancing physical activity (HEPA) remains below the WHO recommendations, and its determinants are not yet fully understood. The study was conducted as part of the international EUPASMOS PLUS project on a sample of [...] Read more.

In Poland—especially in the less developed eastern regions—the level of health-enhancing physical activity (HEPA) remains below the WHO recommendations, and its determinants are not yet fully understood. The study was conducted as part of the international EUPASMOS PLUS project on a sample of 173 adult individuals living in eastern Poland. Physical activity was measured using accelerometers worn continuously for seven days (24/7). The duration of moderate and vigorous physical activity as well as episodes of physical activity lasting at least 10 min were analyzed. The median daily MVPA time was 50 min, and the median VPA time only 10 s, both below WHO recommendations of 150 min/week of MVPA or 75 min/week of VPA. Overall, more than 70% of participants did not meet the recommended levels. The level of HEPA was found to be below WHO recommendations, especially among men, individuals over 50 years old, and those who were professionally inactive. Higher physical activity levels were recorded among women and younger participants. Significant correlations were found between HEPA level and self-rated health status (ρ = 0.28–0.38, p < 0.001), as well as body mass index and waist circumference (ρ ≈ −0.20 to −0.30, p < 0.01). Although statistically significant, the effect sizes were small to moderate, indicating limited explanatory power. Unemployment negatively affected MVPA and VPA levels, while household size positively correlated with physical activity. Interventions promoting HEPA should consider demographic and regional diversity, with particular focus on less active groups such as older adults and the unemployed. It is also necessary to develop new screening tools aimed at easy and quick diagnosis of social groups that should be targeted by HEPA promotion strategies. Full article

(This article belongs to the Special Issue Sensors and Wearable Technologies in Sport Biomechanics—Second Edition)

► Show Figures

Figure 1

32 pages, 4502 KB

Open AccessArticle

An Integrated and Robust Vision System for Internal and External Thread Defect Detection with Adversarial Defense

by Liu Fu, Leqi Li, Gengpei Zhang and Zhihao Jiang

Sensors 2025, 25(18), 5664; https://doi.org/10.3390/s25185664 - 11 Sep 2025

Abstract

In industrial automation, detecting defects in threaded components is challenging due to their complex geometry and the concealment of micro-flaws. This paper presents an integrated vision system capable of inspecting both internal and external threads with high robustness. A unified imaging platform ensures [...] Read more.

In industrial automation, detecting defects in threaded components is challenging due to their complex geometry and the concealment of micro-flaws. This paper presents an integrated vision system capable of inspecting both internal and external threads with high robustness. A unified imaging platform ensures synchronized capture of thread surfaces, while advanced image enhancement techniques improve clarity under motion blur and low-light conditions. To overcome limited defect samples, we introduce a generative data augmentation strategy that diversifies training data. For detection, a lightweight and optimized deep learning model achieves higher precision and efficiency compared with existing YOLO variants. Moreover, we design a dual-defense mechanism that effectively mitigates stealthy adversarial perturbations, such as alpha channel attacks, preserving system reliability. Experimental results demonstrate that the proposed framework delivers accurate, secure, and efficient thread defect detection, offering a practical pathway toward reliable industrial vision systems. Full article

(This article belongs to the Section Intelligent Sensors)

► Show Figures

Figure 1

18 pages, 5871 KB

Open AccessArticle

Inversion of Shear and Longitudinal Acoustic Wave Propagation Parameters in Sea Ice Using SE-ResNet

by Jin Bai, Yi Liu, Xuegang Zhang, Wenmao Yin and Ziye Deng

Sensors 2025, 25(18), 5663; https://doi.org/10.3390/s25185663 - 11 Sep 2025

Abstract

With the advancement of scientific research, understanding the physical parameters governing acoustic wave propagation in sea ice has become increasingly important. Among these parameters, shear wave velocity plays a crucial role. However, as measurements progressed, it became apparent that there was a large [...] Read more.

With the advancement of scientific research, understanding the physical parameters governing acoustic wave propagation in sea ice has become increasingly important. Among these parameters, shear wave velocity plays a crucial role. However, as measurements progressed, it became apparent that there was a large discrepancy between measured values of shear waves and predictions based on empirical formulas or existing models. These inconsistencies stem primarily from the complex internal structure of natural sea ice, which significantly influences its physical behavior. Research reveals that shear wave velocity is not only influenced by bulk properties such as density, temperature, and stress state but is also sensitive to microstructural features, including air bubbles, inclusions, and ice crystal orientation. Compared to longitudinal wave velocity, the characterization of shear wave velocity is far more challenging due to these inherent complexities, underscoring the need for more precise measurement and modeling techniques. To address the challenges posed by the complex internal structure of natural sea ice and improve prediction accuracy, this study introduces a novel, integrated approach combining simulation, measurement, and inversion intelligent learning model. First, a laboratory-based method for generating sea ice layers under controlled formation conditions is developed. The produced sea ice layers align closely with measured values for Poisson’s ratio, multi-year sea ice density, and uniaxial compression modulus, particularly in the high-temperature range. Second, enhancements to shear wave velocity measurement equipment have been implemented. The improved device achieves measurement accuracy exceeding 1%, offers portability, and meets the demands of high-precision experiments conducted in harsh polar environments. Finally, according to the characteristics of small sample data. The ANN neural network was improved to a deep residual neural network with the addition of Squeeze-and-Excitation Attention (SE-ResNet) to predict longitudinal and transverse wave velocities. This prediction method improves the accuracy of shear and longitudinal wave velocity prediction by 24.87% and 39.59%, respectively, compared to the ANN neural network. Full article

(This article belongs to the Topic Recent Advances in Structural Health Monitoring, 2nd Volume)

► Show Figures

Figure 1

23 pages, 2663 KB

Open AccessArticle

Towards Sustainable Personalized Assembly Through Human-Centric Digital Twins

by Marina Crnjac Zizic, Nikola Gjeldum, Marko Mladineo, Bozenko Bilic and Amanda Aljinovic Mestrovic

Sensors 2025, 25(18), 5662; https://doi.org/10.3390/s25185662 - 11 Sep 2025

Abstract

New trends in industry emphasize green and sustainable production on the one hand and personalized or individualized production on the other hand. Introducing new manufacturing technologies and materials to integrate the customer’s specific requirements into the product, while keeping the focus on environmental [...] Read more.

New trends in industry emphasize green and sustainable production on the one hand and personalized or individualized production on the other hand. Introducing new manufacturing technologies and materials to integrate the customer’s specific requirements into the product, while keeping the focus on environmental footprint, becomes a serious challenge. As a result, new production paradigms are developed to keep up with new trends. The most known Industry 4.0 paradigm is oriented towards new technologies and digitalization. Recently, Industry 5.0 appeared as a supplement to the existing Industry 4.0 paradigm, oriented to sustainability and the worker. A multidisciplinary approach is necessary to address these challenges. The Industry 5.0 paradigm’s main pillars—human centricity, resilience, and sustainability—are also pillars of the multidisciplinary approach used in this research. A human-centric approach includes workforce reskilling and acquiring new technologies to ensure that technology serves to enhance human work, while creating a supportive and inclusive work environment and prioritizing employee engagement and wellbeing. Resilience as a second pillar is related to the ability of manufacturing systems and processes to adapt to changing conditions to remain robust and flexible, and sustainability is an important and long-term requirement of this multidisciplinary approach. Based on the research part of the Erasmus+ ExCurS project, particularly research focused on application and training related to digital twins, an advanced concept of organizational sustainability is presented in this paper. The concept of organizational sustainability is realized through the usage of key digital twin technologies aligned with human-centric approaches. A new prototype of a digital twin that optimizes an assembly system based on a developed algorithm and humanoid decision-making is provided as a proof of concept. The human-centric digital twin for industrial application is presented through a case study of personalized products. Full article

(This article belongs to the Special Issue Advances in Intelligent System Engineering for Smart Sensor Applications)

► Show Figures

Figure 1

16 pages, 1755 KB

Open AccessArticle

Development of an Equivalent Circuit to Analyze the Receiving Characteristics of a Class IV Flextensional Transducer

by Eunseo Kang and Yongrae Roh

Sensors 2025, 25(18), 5661; https://doi.org/10.3390/s25185661 - 11 Sep 2025

Abstract

Flextensional transducers are widely utilized as underwater acoustic transducers with broadband and high-sensitivity characteristics in low-frequency ranges. In this study, we have developed an equivalent circuit to facilitate the design of a class IV flextensional hydrophone. The end-plate of the class IV transducer [...] Read more.

Flextensional transducers are widely utilized as underwater acoustic transducers with broadband and high-sensitivity characteristics in low-frequency ranges. In this study, we have developed an equivalent circuit to facilitate the design of a class IV flextensional hydrophone. The end-plate of the class IV transducer is essential for sustaining the structure while keeping the hydrophone waterproof in underwater environments, and the presence or absence of the end-plate changes the hydrophone’s receiving performance. Previous studies have not included the end-plate in their equivalent circuit configurations, but this study proposes a new equivalent circuit model that incorporates an elastic boundary condition representing the constraint imposed on the shell by the compressive force of the end-plate. The receiving voltage sensitivity, calculated using the proposed equivalent circuit model, shows a high degree of agreement with the finite element analysis results, confirming that the mechanical influence of the end-plate significantly affects the hydrophone’s acoustic receiver characteristics. The proposed equivalent circuit approach offers faster computation while maintaining sufficient accuracy compared to the finite element analysis, making it a useful tool for future hydrophone design. Full article

(This article belongs to the Special Issue Feature Papers in Physical Sensors 2025)

► Show Figures

Figure 1

26 pages, 2695 KB

Open AccessArticle

TSN-Interworked Deterministic Transmission over WLAN

by Woojin Ahn

Sensors 2025, 25(18), 5660; https://doi.org/10.3390/s25185660 - 11 Sep 2025

Abstract

Many Time-Sensitive Networking (TSN) workloads require deterministic service across heterogeneous links, yet commodity WLANs are contention-based. Although IEEE 802.11be introduced Restricted Target Wake Time (r-TWT) for prioritized access, its ability to robustly guarantee determinism in mixed deployments with legacy devices remains unverified. We [...] Read more.

Many Time-Sensitive Networking (TSN) workloads require deterministic service across heterogeneous links, yet commodity WLANs are contention-based. Although IEEE 802.11be introduced Restricted Target Wake Time (r-TWT) for prioritized access, its ability to robustly guarantee determinism in mixed deployments with legacy devices remains unverified. We propose a standards-aligned scheme that composes r-TWT, Quiet Time Period (QTP), and an optional Randomized Enqueue (RE) policy. These three mechanisms act in concert to protect the Scheduled Traffic (ST) service period (SP) while minimizing the impact on Non-Scheduled Traffic (NST). To analyze how the proposed scheme impacts existing WLANs, we focus the analysis on how the scheme reshapes the contention period (CP)—where opportunistic capacity is realized—by modeling SP/CP timing with renewal theory and embedding it into an EDCA Markov chain. Simulation results confirm that the proposed scheme protects ST determinism: ST throughput remains pinned to the ceiling with zero observed outage and bounded delay across a wide range of station counts. The proposed scheme minimizes NST throughput degradation in the system-peak throughput range (8–12 stations). Full article

(This article belongs to the Section Sensor Networks)

► Show Figures

Figure 1

17 pages, 5496 KB

Open AccessArticle

Robot-Assisted Mirror Rehabilitation for Post-Stroke Upper Limbs: A Personalized Control Strategy

by Jiayue Chen, Zhongjiang Cheng, Yutong Cai, Shisheng Zhang, Chi Zhu and Yang Zhang

Sensors 2025, 25(18), 5659; https://doi.org/10.3390/s25185659 - 11 Sep 2025

Abstract

To address the limitations of traditional mirror therapy in stroke rehabilitation, such as rigid movement mapping and insufficient personalization, this study proposes a robot-assisted mirror rehabilitation framework integrating multimodal biofeedback. By synchronously capturing kinematic features of the unaffected upper limb and surface electromyography [...] Read more.

To address the limitations of traditional mirror therapy in stroke rehabilitation, such as rigid movement mapping and insufficient personalization, this study proposes a robot-assisted mirror rehabilitation framework integrating multimodal biofeedback. By synchronously capturing kinematic features of the unaffected upper limb and surface electromyography (sEMG) signals from the affected limb, a dual-modal feature fusion network based on a cross-attention mechanism is developed. This network dynamically generates a time-varying mirror ratio coefficient

λ

, which is incorporated into the exoskeleton’s admittance control loop. Combining a trajectory generation algorithm based on dynamic movement primitives (DMPs) with a compliant control strategy incorporating dynamic constraints, the system achieves personalized rehabilitation trajectory planning and safe interaction. Experimental results demonstrate that, compared to traditional mirror therapy, the proposed system exhibits superior performance in bilateral trajectory covariance metrics, the mirror symmetry index, and muscle activation levels. Full article

(This article belongs to the Special Issue New Trends in the Sensing and Control Techniques Used for Intelligent Industrial Perception and Service Robotics)

► Show Figures

Figure 1

31 pages, 41890 KB

Open AccessReview

Comprehensive Review of Open-Source Fundus Image Databases for Diabetic Retinopathy Diagnosis

by Valérian Conquer, Thomas Lambolais, Gustavo Andrade-Miranda and Baptiste Magnier

Sensors 2025, 25(18), 5658; https://doi.org/10.3390/s25185658 - 11 Sep 2025

Abstract

Databases play a crucial role in training, validating, and comparing AI models for detecting retinal diseases, as well as in clinical research, technology development, and healthcare professional training. Diabetic retinopathy (DR), a common diabetes complication, is a leading cause of vision impairment and [...] Read more.

Databases play a crucial role in training, validating, and comparing AI models for detecting retinal diseases, as well as in clinical research, technology development, and healthcare professional training. Diabetic retinopathy (DR), a common diabetes complication, is a leading cause of vision impairment and blindness worldwide. Early detection and management are essential to prevent irreversible vision loss. Fundus photography, known for being economical and non-contact, is a widely applicable gold standard method that offers a convenient way to diagnose and grade DR. This paper presents a comprehensive review of 22 open-source fundus retinal image databases commonly used in DR research, highlighting their main characteristics and key features. Most of these datasets were released between 2000 and 2022. These databases are analyzed through an in-depth examination of their images, enabling objective comparison using color space distances and Principal Component Analysis (PCA) based on 16 key statistical features. Finally, this review aims to support informed decision-making for researchers and practitioners involved in DR diagnosis and management, ultimately improving patient outcomes. Full article

(This article belongs to the Special Issue Digital Image Processing and Sensing Technologies—Second Edition)

► Show Figures

Figure 1

15 pages, 473 KB

Open AccessArticle

Every Step Counts—How Can We Accurately Count Steps with Wearable Sensors During Activities of Daily Living in Individuals with Neurological Conditions?

by Florence Crozat, Johannes Pohl, Chris Easthope Awai, Christoph Michael Bauer and Roman Peter Kuster

Sensors 2025, 25(18), 5657; https://doi.org/10.3390/s25185657 - 11 Sep 2025

Abstract

Wearable sensors provide objective, continuous, and non-invasive quantification of physical activity, with step count serving as one of the most intuitive measures. However, significant gait alterations in individuals with neurological conditions limit the accuracy of step-counting algorithms trained on able-bodied individuals. Therefore, this [...] Read more.

Wearable sensors provide objective, continuous, and non-invasive quantification of physical activity, with step count serving as one of the most intuitive measures. However, significant gait alterations in individuals with neurological conditions limit the accuracy of step-counting algorithms trained on able-bodied individuals. Therefore, this study investigates the accuracy of step counting during activities of daily living (ADL) in a neurological population. Seven individuals with neurological conditions wore seven accelerometers while performing ADL for 30 min. Step events manually annotated from video served as ground truth. An optimal sensing and analysis configuration for machine learning algorithm development (sensor location, filter range, window length, and regressor type) was identified and compared to existing algorithms developed for able-bodied individuals. The most accurate configuration includes a waist-worn sensor, a 0.5–3 Hz bandpass filter, a 5 s window, and gradient boosting regression. The corresponding algorithm showed a significantly lower error rate compared to existing algorithms trained on able-bodied data. Notably, all algorithms undercounted steps. This study identified an optimal sensing and analysis configuration for machine learning-based step counting in a neurological population and highlights the limitations of applying able-bodied-trained algorithms. Future research should focus on developing accurate and robust step-counting algorithms tailored to individuals with neurological conditions. Full article

(This article belongs to the Section Wearables)

► Show Figures

Graphical abstract

18 pages, 3542 KB

Open AccessArticle

Design and Implementation of a Cascade Control System for a Variable Air Volume in Operating Rooms Based on Pressure and Temperature Feedback

by Abdulmohaymin Bassim Qassim, Shaimaa Mudhafar Hashim and Wajdi Sadik Aboud

Sensors 2025, 25(18), 5656; https://doi.org/10.3390/s25185656 - 10 Sep 2025

Abstract

This research presents the design and implementation of a cascade Proportional–Integral (PI) controller tailored for a Variable Air Volume (VAV) system that was specially created and executed particularly for hospital operating rooms. The main goal of this work is to make sure that [...] Read more.

This research presents the design and implementation of a cascade Proportional–Integral (PI) controller tailored for a Variable Air Volume (VAV) system that was specially created and executed particularly for hospital operating rooms. The main goal of this work is to make sure that the temperature and positive pressure stay within the limits set by ASHRAE Standard 170-2017. This is necessary for patient safety, surgical accuracy, and system reliability. The proposed cascade design uses dual-loop PI controllers: one loop controls the temperature based on user-defined setpoints by local control touch screen, and the other loop accurately modulates the differential pressure to keep the pressure of the environment sterile (positive pressure). The system works perfectly with Building Automation System (BAS) parts from Automated Logic Corporation (ALC) brand, like Direct Digital Controllers (DDC) and Web-CTRL software with Variable Frequency Drives (VFDs), advanced sensors, and actuators that give real-time feedback, precise control, and energy efficiency. The system’s exceptional responsiveness, extraordinary stability, and resilient flexibility were proven through empirical validation at the Korean Iraqi Critical Care Hospital in Baghdad under a variety of operating circumstances. Even during rapid load changes and door openings, the control system successfully maintained the temperature between 18 and 22 °C and the differential pressure between 3 and 15 Pascals. Four performance scenarios, such as normal (pressure and temperature), high-temperature, high-pressure, and low-pressure cases, were tested. The results showed that the cascade PI control strategy is a reliable solution for critical care settings because it achieves precise environmental control, improves energy efficiency, and ensures compliance with strict healthcare facility standards. Full article

(This article belongs to the Section Industrial Sensors)

► Show Figures

Figure 1

30 pages, 5137 KB

Open AccessArticle

High-Resolution Remote Sensing Imagery Water Body Extraction Using a U-Net with Cross-Layer Multi-Scale Attention Fusion

by Chunyan Huang, Mingyang Wang, Zichao Zhu and Yanling Li

Sensors 2025, 25(18), 5655; https://doi.org/10.3390/s25185655 - 10 Sep 2025

Abstract

The accurate extraction of water bodies from remote sensing imagery is crucial for water resource monitoring and flood disaster warning. However, this task faces significant challenges due to complex land cover, large variations in water body morphology and spatial scales, and spectral similarities [...] Read more.

The accurate extraction of water bodies from remote sensing imagery is crucial for water resource monitoring and flood disaster warning. However, this task faces significant challenges due to complex land cover, large variations in water body morphology and spatial scales, and spectral similarities between water and non-water features, leading to misclassification and low accuracy. While deep learning-based methods have become a research hotspot, traditional convolutional neural networks (CNNs) struggle to represent multi-scale features and capture global water body information effectively. To enhance water feature recognition and precisely delineate water boundaries, we propose the AMU-Net model. Initially, an improved residual connection module was embedded into the U-Net backbone to enhance complex feature learning. Subsequently, a multi-scale attention mechanism was introduced, combining grouped channel attention with multi-scale convolutional strategies for lightweight yet precise segmentation. Thereafter, a dual-attention gated modulation module dynamically fusing channel and spatial attention was employed to strengthen boundary localization. Furthermore, a cross-layer geometric attention fusion module, incorporating grouped projection convolution and a triple-level geometric attention mechanism, optimizes segmentation accuracy and boundary quality. Finally, a triple-constraint loss framework synergistically optimized global classification, regional overlap, and background specificity to boost segmentation performance. Evaluated on the GID and WHDLD datasets, AMU-Net achieved remarkable IoU scores of 93.6% and 95.02%, respectively, providing an effective new solution for remote sensing water body extraction. Full article

(This article belongs to the Section Remote Sensors)

► Show Figures

Figure 1

23 pages, 4599 KB

Open AccessReview

In Vitro Evaluation of Confounders in Brain Optical Monitoring: A Review

by Karina Awad-Pérez, Maria Roldan and Panicos A. Kyriacou

Sensors 2025, 25(18), 5654; https://doi.org/10.3390/s25185654 - 10 Sep 2025

Abstract

Optical brain monitoring techniques, including near-infrared spectroscopy (NIRS), diffuse correlation spectroscopy (DCS), and photoplethysmography (PPG) have gained attention for their non-invasive, affordable, and portable nature. These methods offer real-time insights into cerebral parameters like cerebral blood flow (CBF), intracranial pressure (ICP), and oxygenation. [...] Read more.

Optical brain monitoring techniques, including near-infrared spectroscopy (NIRS), diffuse correlation spectroscopy (DCS), and photoplethysmography (PPG) have gained attention for their non-invasive, affordable, and portable nature. These methods offer real-time insights into cerebral parameters like cerebral blood flow (CBF), intracranial pressure (ICP), and oxygenation. However, confounding factors like extracerebral layers, skin pigmentation, skull thickness, and brain-related pathologies may affect measurement accuracy. This review examines the potential impact of confounders, focusing on in vitro studies that use phantoms to simulate human head properties under controlled conditions. A systematic search identified six studies on extracerebral layers, two on skin pigmentation, two on skull thickness, and four on brain pathologies. While variation in phantom designs and optical devices limits comparability, findings suggest that the extracerebral layer and skull thickness influence measurement accuracy, and skin pigmentation introduces bias. Pathologies like oedema and haematomas affect the optical signal, though their influence on parameter estimation remains inconclusive. This review highlights limitations in current research and identifies areas for future investigation, including the need for improved brain phantoms capable of simulating pulsatile signals to assess the impact of confounders on PPG systems, given the growing interest in PPG-based cerebral monitoring. Addressing these challenges will improve the reliability of optical monitoring technologies. Full article

(This article belongs to the Collection Sensors for Globalized Healthy Living and Wellbeing)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Sensors, Volume 25, Issue 18 (September-2 2025) – 76 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI