Sensors

14 pages, 3710 KiB

Open AccessArticle

An Extrinsic Optical Fiber Sensor Probe with Micrometer Size via a C-Shaped Waveguide with a Core of MIP

by Chiara Marzano, Rosalba Pitruzzella, Francesco Arcadio, Filipa Sequeira, Luca Pasquale Renzullo, Alessandra Cutaia, Catarina Cardoso Novo, Ricardo Oliveira, Maria Pesavento, Luigi Zeni, Giancarla Alberti, Nunzio Cennamo and Rogerio Nunes Nogueira

Sensors 2025, 25(10), 3250; https://doi.org/10.3390/s25103250 - 21 May 2025

Viewed by 151

Abstract

Optical–chemical sensors based on optical fibers can be made in reflection or transmission schemes. In the reflection scheme, the sensing area is typically present at the end of the fiber, and the light source and the detector are placed on the same side [...] Read more.

Optical–chemical sensors based on optical fibers can be made in reflection or transmission schemes. In the reflection scheme, the sensing area is typically present at the end of the fiber, and the light source and the detector are placed on the same side of the fiber. This approach can be exploited to achieve chemical probes useful in several application fields where remote sensing is required. In this work, to obtain an extrinsic optical fiber chemical sensor in a reflection scheme, two optical fibers are used to monitor a chemically sensitive region achieved by a C-shaped waveguide with a molecularly imprinted polymer (MIP) as a core between the optical fibers. The proposed micrometer-sized probe is developed and tested as a proof of concept via a MIP for 2-Furaldehyde (2-FAL) detection of interest in food and industrial applications. The experimental results of the proposed sensing approach showed several advantages, such as a nanomolar detection limit and an ultra-wide concentration detection range due to different kinds of MIP recognition sites in the optical path between the fibers. Full article

(This article belongs to the Section Optical Sensors)

► Show Figures

Figure 1

21 pages, 922 KiB

Open AccessArticle

DBN-BAAE: Enhanced Lightweight Anomaly Detection Mechanism with Boosting Adversarial Autoencoder

by Yanru Chen, Bei Wu, Wang Zhong, Yanru Guo, Dizhi Wu, Yi Ren and Yuanyuan Zhang

Sensors 2025, 25(10), 3249; https://doi.org/10.3390/s25103249 - 21 May 2025

Viewed by 117

Abstract

The growing digitalization of Industrial Control Systems (ICSs) presents both significant benefits and security challenges, especially for small and medium-sized factories with limited resources. Effective anomaly detection is essential to safeguard these facilities and prevent costly disruptions. Although current research has advanced anomaly [...] Read more.

The growing digitalization of Industrial Control Systems (ICSs) presents both significant benefits and security challenges, especially for small and medium-sized factories with limited resources. Effective anomaly detection is essential to safeguard these facilities and prevent costly disruptions. Although current research has advanced anomaly detection, it is still challenging for algorithms to be capable of effectively balancing the interplay between training speed, computational cost, and accuracy while simultaneously exhibiting robust stability and adaptability. This gap often leaves small and medium-sized factories without efficient solutions. To address these issues, this work introduces a deep belief network-based boosting adversarial autoencoder termed DBN-BAAE, a novel lightweight anomaly detection mechanism based on boosting adversarial learning. The proposed lightweight mechanism saves computational overhead, enhances autoencoder training stability with an improved deep belief network (DBN) for pre-training, boosts encoder expression through ensemble learning, achieves high detection accuracy via an adversarial decoder, and employs a dynamic threshold to enhance adaptability and reduce the need for retraining. Experiments reveal that the mechanism not only achieves an F1 score of 0.82, surpassing the best baseline by 1%, but also accelerates training speed by 2.2 times, demonstrating its effectiveness and efficiency in ICS environments, particularly for small and medium-sized factories. Full article

(This article belongs to the Topic Artificial Intelligence in Smart Industrial Diagnostics and Manufacturing—Third Edition)

► Show Figures

Figure 1

27 pages, 1101 KiB

Open AccessArticle

Microservice Deployment Based on Multiple Controllers for User Response Time Reduction in Edge-Native Computing

by Zhaoyang Wang, Jinqi Zhu, Jia Guo and Yang Liu

Sensors 2025, 25(10), 3248; https://doi.org/10.3390/s25103248 - 21 May 2025

Viewed by 139

Abstract

Microservice deployment methods in edge-native computing environments hold great potential for minimizing user application response time. However, most existing studies overlook the communication overhead between microservices and controllers, as well as the impact of microservice pull time on user response time. To address [...] Read more.

Microservice deployment methods in edge-native computing environments hold great potential for minimizing user application response time. However, most existing studies overlook the communication overhead between microservices and controllers, as well as the impact of microservice pull time on user response time. To address these issues, this paper proposes a multi-controller service mesh architecture to reduce data transfer overhead between microservices and controllers. Furthermore, we formulate the microservice deployment problem as an optimization problem aimed at minimizing both system communication overhead and microservice deployment cost. To achieve this, we introduce a novel RIME optimization algorithm and enhanced Adaptive Crested Porcupine Optimizer (RIME-ACPO) algorithm that optimizes microservice placement decisions. Notably, this algorithm incorporates a real-time resource monitoring-based load balancing algorithm, dynamically adjusting microservice deployment according to edge server resource utilization to enhance the execution performance of user applications. Finally, extensive simulation experiments were conducted to validate the effectiveness of the proposed algorithm. The experimental results demonstrate that, compared with other algorithms, our algorithm significantly improves user response time, optimizes resource utilization, and reduces the total cost. Full article

(This article belongs to the Section Internet of Things)

► Show Figures

Figure 1

20 pages, 4838 KiB

Open AccessArticle

Assessment of RF Electromagnetic Exposure to Car Driver from Monopole Array Antennas in V2V Communications Considering Thermal Characteristics

by Shirun Wang and Mai Lu

Sensors 2025, 25(10), 3247; https://doi.org/10.3390/s25103247 - 21 May 2025

Viewed by 116

Abstract

Vehicles are rapidly evolving into objects of intelligent interconnection. Vehicle-to-Vehicle (V2V) communications enable the interconnection between vehicles, while also leading to new electromagnetic exposure scenarios. This paper integrates a monopole array antenna into a shark-fin antenna on the car roof for V2V communications [...] Read more.

Vehicles are rapidly evolving into objects of intelligent interconnection. Vehicle-to-Vehicle (V2V) communications enable the interconnection between vehicles, while also leading to new electromagnetic exposure scenarios. This paper integrates a monopole array antenna into a shark-fin antenna on the car roof for V2V communications and evaluates the specific absorption rate (SAR) and temperature rise of a human body in a smart mobility communication scenario operating at 5.9 GHz. The V2V antenna is modeled and placed on a 3D vehicle model using COMSOL Multiphysics (v.6.2) to numerically estimate the SAR in the head and body regions of the human body model (adult male) inside the vehicle. Both the localized and whole-body 30 min average SAR are lower than the International Commission on Non-Ionizing Radiation Protection (ICNIRP) occupational restrictions for electromagnetic field exposure from 100 kHz to 6 GHz, being equal in the worst-case scenario to 0.981 W/kg (for the head), which is 9.81% of the ICNIRP limit (10 W/kg), and 0.008728 W/kg (for the whole-body average), which is 2.18% of the ICNIRP limit (0.4 W/kg). The 30 min average human core temperature rise is 0.055 °C, which is 5.5% of the ICNIRP limit. This indicates that, in typical automotive scenarios, the electromagnetic exposure from a monopole array antenna for V2V communications does not pose threat to the human body. This study provides knowledge related to emerging exposure scenarios in intelligent mobility communication, which is beneficial for evaluating possible health impacts and designing public health management policies. Full article

(This article belongs to the Section Vehicular Sensing)

► Show Figures

Figure 1

23 pages, 6489 KiB

Open AccessArticle

Removing Random Noise of GPR Data Using Joint BM3D−IAM Filtering

by Wentian Wang, Wei Du and Zhuo Jia

Sensors 2025, 25(10), 3246; https://doi.org/10.3390/s25103246 - 21 May 2025

Viewed by 139

Abstract

Random noise degrades the quality and reduces the interpretability of Ground Penetrating Radar (GPR) data. The Block Matching Three Dimension (BM3D) algorithm is effective at suppressing Gaussian noise, but ineffective at handling salt-and-pepper noise. On the other hand, the Improved Adaptive Median (IAM) [...] Read more.

Random noise degrades the quality and reduces the interpretability of Ground Penetrating Radar (GPR) data. The Block Matching Three Dimension (BM3D) algorithm is effective at suppressing Gaussian noise, but ineffective at handling salt-and-pepper noise. On the other hand, the Improved Adaptive Median (IAM) filter is suitable for eliminating salt-and-pepper noise, but performs poorly against Gaussian noise. In this paper, we introduce and implement JBI, a joint denoising algorithm that integrates both BM3D and improved adaptive median filtering, exploiting the advantages of both algorithms to effectively remove both Gaussian and salt-and-pepper noise from GPR data. Applying the proposed joint filter to both synthetic and real field GPR data, infested with various proportions of different noise types, shows that the proposed joint denoising algorithm yields significantly better results than these two filters when used separately, and better than other commonly used denoising filters. Full article

(This article belongs to the Special Issue Radars, Sensors and Applications for Applied Geophysics)

► Show Figures

Figure 1

17 pages, 5479 KiB

Open AccessArticle

On the Effect of Layering Velostat on Force Sensing for Hands

by Tyler Bartunek, Ann Majewicz Fey and Edoardo Battaglia

Sensors 2025, 25(10), 3245; https://doi.org/10.3390/s25103245 - 21 May 2025

Viewed by 95

Abstract

Force sensing on hands can provide an understanding of interaction forces during manipulation, with applications in different fields, including robotics and medicine. While several approaches to accomplish this have been proposed, they often require relatively complex and/or expensive fabrication techniques and materials. On [...] Read more.

Force sensing on hands can provide an understanding of interaction forces during manipulation, with applications in different fields, including robotics and medicine. While several approaches to accomplish this have been proposed, they often require relatively complex and/or expensive fabrication techniques and materials. On the other hand, less complex and expensive approaches often suffer from poor accuracy of measurements. An example of this is provided by sensors built with Velostat, a polyethylene–carbon composite material that exhibits resistance changes when force is applied. This material is both cheap and easy to work with, but sensors made from Velostat have been shown to suffer from low accuracy, limiting its usefulness. This work explores the effect of stacking multiple layers of 0.1 mm Velostat sheets on accuracy, using no additional fabrication techniques or other material aside from electrode connections, with the rationale that this is both economical and can be accomplished easily. We evaluate measurement error for designs with different numbers of layers (1, 3, 4, 5, 10, 20, and 30) against a load cell, and also compare this with the error for a USD 10 commercial force sensing resistor designed for measurement of hand forces (FSR 402) in three evaluations (static, cyclic, and finger base interactions). Our results show that layered sensors outperform both the one-layer design and the commercial FSR sensor consistently under all conditions considered, with the best performing sensors reducing measurement errors by at least 27% and as much as 60% when compared against the one-layer design. Full article

(This article belongs to the Special Issue Flexible Pressure/Force Sensors and Their Applications)

► Show Figures

Graphical abstract

16 pages, 570 KiB

Open AccessCommunication

Multiple-Cumulant-Matrices-Based Method for Exact NF Polarization Localization with COLD Array

by Jiefeng Zheng, Haifen Meng, Zhuang Luo, Huayue Wu, Weiyue Liu and Hua Chen

Sensors 2025, 25(10), 3244; https://doi.org/10.3390/s25103244 - 21 May 2025

Viewed by 80

Abstract

As a key technology for the fifth-generation of mobile communications, massive MIMO systems enable massive user access via large-scale arrays. However, their dense deployment extends the near-field (NF) region, introducing new localization complexities. Based on an exact spherical wavefront model, this paper proposes [...] Read more.

As a key technology for the fifth-generation of mobile communications, massive MIMO systems enable massive user access via large-scale arrays. However, their dense deployment extends the near-field (NF) region, introducing new localization complexities. Based on an exact spherical wavefront model, this paper proposes a multiple-cumulant-matrices-based method for NF source localization using a Co-centered Orthogonal Loop and Dipole (COLD) array. Firstly, following the physical numbering of array elements, we can construct multiple polarization cumulant matrices, which can then be cascaded into a long matrix. Next, the signal subspace can be obtained through eigen-decomposition of this long matrix, from which the horizontal and vertical components can be further separated. By applying ESPRIT, joint angle, range, and polarization parameters can be estimated. In addition, the asymptotic variances for joint spatial and polarization parameters are analyzed. Compared with existing NF polarization algorithms, the proposed method exhibits better parameter estimation and is consistent with a theoretical asymptotic performance. Full article

(This article belongs to the Special Issue Recent Advances in Sensor Array Signal Processing and its Applications in Future Communication and Radar)

► Show Figures

Figure 1

22 pages, 3511 KiB

Open AccessArticle

An Investigation of Real-Time Galileo/GPS Integrated Precise Kinematic Time Transfer Based on Galileo HAS Service

by Lei Xu, Shaoxin Chen, Yuanyuan An, Pengli Shen, Xia Xiao, Qianqian Chen, Jianxiong Wei, Yao Chen and Ye Yu

Sensors 2025, 25(10), 3243; https://doi.org/10.3390/s25103243 - 21 May 2025

Viewed by 104

Abstract

GNSS Precise Point Positioning (PPP) technology has been extensively applied to post-processing international comparisons between UTC/TAI times and real-time time transfer, predominantly in static configurations. However, with the swift advancement of intelligent and unmanned systems, there is an urgent need for research into [...] Read more.

GNSS Precise Point Positioning (PPP) technology has been extensively applied to post-processing international comparisons between UTC/TAI times and real-time time transfer, predominantly in static configurations. However, with the swift advancement of intelligent and unmanned systems, there is an urgent need for research into kinematic time transfer. This paper introduces a kinematic model Galileo/GPS integrated PPP time transfer approach leveraging the Galileo High Accuracy Service (HAS). The study utilized observational data from seven stations spanning 22 days. The findings indicate that under static conditions, GPS, Galileo, and Galileo/GPS PPP, when supported by the Galileo HAS, can achieve time transfer with sub-nanosecond precision. In kinematic scenarios, the accuracy of single-system PPP time transfer is comparatively lower, with frequent re-convergence events leading to significant accuracy degradation (exceeding 1 ns). However, in cases where re-convergence is infrequent due to a limited number of satellites, sub-nanosecond time transfer is still attainable. The Galileo/GPS integrated PPP time transfer effectively mitigates the issue of re-convergence, ensuring sub-nanosecond accuracy across all links (0.48 ns). Consequently, it is recommended to employ a multi-system integration approach for kinematic PPP time transfer, particularly when utilizing the Galileo HAS. In terms of frequency stability, GPS, Galileo, and Galileo/GPS PPP demonstrate short-term stability (over 960 s) of (5.29 × 10⁻¹³, 3.34 × 10⁻¹³, and 1.60 × 10⁻¹³), respectively, and long-term stability (over 15,360 s) of (1.49 × 10⁻¹³, 1.02 × 10⁻¹³, and 4.06 × 10⁻¹⁴), respectively. Full article

(This article belongs to the Special Issue Advancements in GNSS Precise Point Positioning Technology and Applications)

► Show Figures

Figure 1

23 pages, 1739 KiB

Open AccessArticle

Design of a Lorentz Force Magnetic Bearing Group Steering Law Based on an Adaptive Weighted Pseudo-Inverse Law

by Chenyu Wang, Lei Li, Weijie Wang, Yanbin Zhao, Baiqi Li and Yuan Ren

Sensors 2025, 25(10), 3242; https://doi.org/10.3390/s25103242 - 21 May 2025

Viewed by 84

Abstract

Aiming at the high-precision torque output and saturation singularity avoidance problems in Lorentz force magnetic bearing (LFMB) swarms for magnetic levitation spacecraft, this study designs a manipulation law based on an adaptive weighted pseudo-inverse law. The system monitors each magnetic bearing’s working state [...] Read more.

Aiming at the high-precision torque output and saturation singularity avoidance problems in Lorentz force magnetic bearing (LFMB) swarms for magnetic levitation spacecraft, this study designs a manipulation law based on an adaptive weighted pseudo-inverse law. The system monitors each magnetic bearing’s working state in real time using high-precision position and current sensors. As the key input for the adaptive weighted pseudo-inverse control law, the sensor data’s measurement accuracy directly determines torque distribution effectiveness and attitude control precision. First, considering electromagnetic back-EMF effects, individual LFMB dynamics are modeled via the equivalent magnetic circuit method, with working principles elucidated. Subsequently, saturation coefficients for LFMB swarms are designed. Incorporating spacecraft maneuvering requirements, a genetic optimization algorithm establishes the optimal mounting configuration under task constraints. Considering the LFMB swarm configuration characteristics, this study proposes an adaptive weighted pseudo-inverse maneuvering law tailored to operational constraints. By designing an adaptive weighting matrix, the maneuvering law adjusts each LFMB’s torque output in real time, reducing residual saturation effects on attitude control speed and accuracy. Simulation results demonstrate that the proposed mounting configuration and adaptive weighted pseudo-inverse maneuvering law effectively mitigate saturation singularity’s impact on attitude control accuracy while reducing total energy consumption by 22%, validating the method’s effectiveness and superiority. Full article

(This article belongs to the Special Issue Magnetic Sensors and Systems for Scientific and Industrial Applications)

► Show Figures

Figure 1

26 pages, 3350 KiB

Open AccessArticle

Optimizing Backbone Networks Through Hybrid–Modal Fusion: A New Strategy for Waste Classification

by Houkui Zhou, Qifeng Ding, Chang Chen, Qinqin Liao, Qun Wang, Huimin Yu, Haoji Hu, Guangqun Zhang, Junguo Hu and Tao He

Sensors 2025, 25(10), 3241; https://doi.org/10.3390/s25103241 - 21 May 2025

Viewed by 154

Abstract

With rapid urbanization, effective waste classification is a critical challenge. Traditional manual methods are time-consuming, labor-intensive, costly, and error-prone, resulting in reduced accuracy. Deep learning has revolutionized this field. Convolutional neural networks such as VGG and ResNet have dramatically improved automated sorting efficiency, [...] Read more.

With rapid urbanization, effective waste classification is a critical challenge. Traditional manual methods are time-consuming, labor-intensive, costly, and error-prone, resulting in reduced accuracy. Deep learning has revolutionized this field. Convolutional neural networks such as VGG and ResNet have dramatically improved automated sorting efficiency, and Transformer architectures like the Swin Transformer have further enhanced performance and adaptability in complex sorting scenarios. However, these approaches still struggle in complex environments and with diverse waste types, often suffering from limited recognition accuracy, poor generalization, or prohibitive computational demands. To overcome these challenges, we propose an efficient hybrid-modal fusion method, the Hybrid-modal Fusion Waste Classification Network (HFWC-Net), for precise waste image classification. HFWC-Net leverages a Transformer-based hierarchical architecture that integrates CNNs and Transformers, enhancing feature capture and fusion across varied image types for superior scalability and flexibility. By incorporating advanced techniques such as the Agent Attention mechanism and the LionBatch optimization strategy, HFWC-Net not only improves classification accuracy but also significantly reduces classification time. Comparative experimental results on the public datasets Garbage Classification, TrashNet, and our self-built MixTrash dataset demonstrate that HFWC-Net achieves Top-1 accuracy rates of 98.89%, 96.88%, and 94.35%, respectively. These findings indicate that HFWC-Net attains the highest accuracy among current methods, offering significant advantages in accelerating classification efficiency and supporting automated waste management applications. Full article

(This article belongs to the Special Issue Artificial Intelligence-Based Target Recognition and Remote Sensing Data Processing)

► Show Figures

Figure 1

13 pages, 5874 KiB

Open AccessArticle

Fano Resonance Mach–Zehnder Modulator Based on a Single Arm Coupled with a Photonic Crystal Nanobeam Cavity for Silicon Photonics

by Enze Shi, Guang Chen, Lidan Lu, Yingjie Xu, Jieyu Yang and Lianqing Zhu

Sensors 2025, 25(10), 3240; https://doi.org/10.3390/s25103240 - 21 May 2025

Viewed by 169

Abstract

Recently, Fano resonance modulators and photonic crystal nanobeam cavities (PCNCs) have attracted more and more attention due to their superior performance, such as high modulation efficiency and high extinction ratio (ER). In this paper, a silicon Fano resonance Mach–Zehnder modulator (MZM) based on [...] Read more.

Recently, Fano resonance modulators and photonic crystal nanobeam cavities (PCNCs) have attracted more and more attention due to their superior performance, such as high modulation efficiency and high extinction ratio (ER). In this paper, a silicon Fano resonance Mach–Zehnder modulator (MZM) based on a single arm coupled with a PCNC is theoretically analyzed, designed, and numerically simulated. By optimizing the coupling length, lattice constant, coupling gap, and the number of holes in the mirror/taper region, the ER of our MZM can achieve 34 dB. When the applied voltage of the MZM is biased at 4.3 V and the non-return-to-zero on–off keying (NRZ-OOK) signal at a data rate of 10 Gbit/s is modulated, the sharpest asymmetric resonant peak and the most remarkable Fano line shape can be obtained around a wavelength of 1550.68 nm. Compared with the traditional nanobeam cavities, along with the varying radii, our PCNC design has holes with a fixed radius of 90 nm, which is suitable to be fabricated by a 180 nm passive silicon photonic multi-project wafer (MPW). Therefore, our compacted lab-on-chip, resonance-based silicon photonic MZM that is coupled with a PCNC has the advantages of superior performance and easy fabrication, which provide support for photonic integrated circuit designs and can be beneficial to various silicon photonic application fields, including photonic computing, photonic convolutional neural networks, and optical communications, in the future. Full article

(This article belongs to the Special Issue Advances in Microwave Photonics)

► Show Figures

Figure 1

25 pages, 2185 KiB

Open AccessArticle

Analytical Framework for Online Calibration of Sensor Systematic Errors Under the Generic Multisensor Integration Strategy

by Benjamin Brunson and Jianguo Wang

Sensors 2025, 25(10), 3239; https://doi.org/10.3390/s25103239 - 21 May 2025

Viewed by 88

Abstract

This paper proposes an analytical framework for pre-analyzing the potential performance of online sensor calibration in Kalman filtering. Taking a multi-sensor integrated kinematic positioning and navigation system as an example, a pre-analysis of the system performance can be conducted: the observability of individual [...] Read more.

This paper proposes an analytical framework for pre-analyzing the potential performance of online sensor calibration in Kalman filtering. Taking a multi-sensor integrated kinematic positioning and navigation system as an example, a pre-analysis of the system performance can be conducted: the observability of individual sensor systematic error states; minimum estimable values of sensor systematic error states; and minimum detectable systematic errors in sensor observations. These measures together allow for a rigorous characterization of the potential performance of a system as part of mission planning. The proposed framework enables a thorough evaluation of the relative value of different calibration maneuvers and sensor configurations before data collection by simulating the anticipated trajectory, without even requiring the construction of a physical system. When used with the Generic Multisensor Integration Strategy (GMIS), the proposed framework provides unique insight into the potential performance of IMU sensors. To illustrate the utility of the proposed framework, two situations were analyzed: one where no specific calibration maneuvers were undertaken and one where a circular motion maneuver was undertaken. The results show the potential and practicality of the proposed framework in firmly establishing best practices for field procedures and learning about the system’s capability when using online sensor calibration. Full article

(This article belongs to the Section Intelligent Sensors)

► Show Figures

Figure 1

16 pages, 534 KiB

Open AccessSystematic Review

Evaluating the Clinical Utility of Robotic Systems in Plastic and Reconstructive Surgery: A Systematic Review

by Ishith Seth, Kaiyang Lim, Edmond Chang, Warren M. Rozen and Sally Kiu-Huen Ng

Sensors 2025, 25(10), 3238; https://doi.org/10.3390/s25103238 - 21 May 2025

Viewed by 125

Abstract

Background: Robotic surgical systems offer enhanced precision, motion scaling, tremor filtration, and visualization, making them highly suitable for the complex anatomical demands of plastic and reconstructive surgery. While widely implemented in other specialties, their integration in plastic surgery remains limited. This systematic [...] Read more.

Background: Robotic surgical systems offer enhanced precision, motion scaling, tremor filtration, and visualization, making them highly suitable for the complex anatomical demands of plastic and reconstructive surgery. While widely implemented in other specialties, their integration in plastic surgery remains limited. This systematic review evaluates the clinical applications, outcomes, and limitations of robotic-assisted techniques in plastic and reconstructive procedures. Methods: Following PRISMA guidelines, a systematic search was conducted across PubMed, Embase, Scopus, and Web of Science for studies published between January 1980 and March 2025. Clinical studies reporting robotic applications in plastic surgery were included, while cadaveric, animal, and non-English studies were excluded. Data extraction and quality assessment were performed using Covidence and validated tools including the CARE checklist, NOS, GRADE, and SYRCLE. A total of 1428 studies were screened, and 31 met the inclusion criteria. Results: Robotic systems were primarily applied in microsurgery (n = 16), breast reconstruction (n = 8), and craniofacial/aesthetic surgery (n = 7). Common platforms included the Symani Surgical System, Da Vinci systems, and ARTAS. Robotic-assisted approaches improved precision, aesthetic outcomes, flap survival, and patient satisfaction, particularly in procedures involving lymphaticovenous anastomosis and nipple-sparing mastectomy. However, challenges included steep learning curves, longer operative times, high equipment costs, and the lack of haptic feedback. Quality assessment rated all studies as moderate. Conclusions: Robotic-assisted surgery demonstrates considerable potential in enhancing plastic and reconstructive outcomes. As systems become more compact, cost-effective, and integrated with AI and biomimetic technologies, their broader adoption is anticipated. Further high-quality studies are needed to optimize these systems and support widespread clinical implementation. Full article

(This article belongs to the Special Issue Advance in Sensors Technology for Medical Robotics)

► Show Figures

Figure 1

17 pages, 3268 KiB

Open AccessArticle

Simulative Analysis of Stimulated Raman Scattering Effects on WDM-PON Based 5G Fronthaul Networks

by Yan Xu, Shuai Wang and Asad Saleem

Sensors 2025, 25(10), 3237; https://doi.org/10.3390/s25103237 - 21 May 2025

Viewed by 91

Abstract

In future hybrid fiber and radio access networks, wavelength division multiplexing passive optical networks (WDM-PON) based fifth-generation (5G) fronthaul systems are anticipated to coexist with current protocols, potentially leading to non-linearity impairment due to stimulated Raman scattering (SRS). To meet the loss budget [...] Read more.

In future hybrid fiber and radio access networks, wavelength division multiplexing passive optical networks (WDM-PON) based fifth-generation (5G) fronthaul systems are anticipated to coexist with current protocols, potentially leading to non-linearity impairment due to stimulated Raman scattering (SRS). To meet the loss budget requirements of 5G fronthaul networks, this paper investigates the power changes induced by SRS in WDM-PON based 5G fronthaul systems. The study examines wavelength allocation schemes utilizing both the C-band and O-band, with modulation formats including non-return-to-zero (NRZ), optical double-binary (ODB), and four-level pulse amplitude modulation (PAM4). Simulation results indicate that SRS non-linearity impairment causes a power depletion of 1.3 dB in the 20 km C-band link scenario, regardless of whether the modulation formats are 25 Gb/s or 50 Gb/s NRZ, ODB, and PAM4, indicating that the SRS-induced power changes are largely independent of both modulation formats and modulation rates. This effect occurs when only the upstream and downstream wavelengths of the 5G fronthaul are broadcast. However, when the 5G fronthaul wavelengths coexist with previous protocols, the maximum power depletion increases significantly to 10.1 dB. In the O-band scenario, the SRS-induced maximum power depletion reaches 1.5 dB with NRZ, ODB, and PAM4 modulation formats at both 25 Gb/s and 50 Gb/s. Based on these analyses, the SRS non-linearity impairment shall be fully considered when planning the wavelengths for 5G fronthaul transmission. Full article

(This article belongs to the Special Issue Novel Technology in Optical Communications)

► Show Figures

Figure 1

16 pages, 1402 KiB

Open AccessArticle

Fault Diagnosis of Switching Power Supplies Using Dynamic Wavelet Packet Transform and Optimized SVM

by Jie Xu, Jingjing Zhu and Zhifeng Wang

Sensors 2025, 25(10), 3236; https://doi.org/10.3390/s25103236 - 21 May 2025

Viewed by 118

Abstract

Switch mode power supplies (SMPSs) are prone to various faults under complex operating environments and variable load conditions. To improve the accuracy and reliability of fault diagnosis, this paper proposes an intelligent diagnosis method based on Dynamic Wavelet Packet Transform (DWPT) and Improved [...] Read more.

Switch mode power supplies (SMPSs) are prone to various faults under complex operating environments and variable load conditions. To improve the accuracy and reliability of fault diagnosis, this paper proposes an intelligent diagnosis method based on Dynamic Wavelet Packet Transform (DWPT) and Improved Artificial Bee Colony Optimized Support Vector Machine (APABC-SVM). First, an adaptive wavelet packet decomposition mechanism is used to refine the time–frequency feature extraction of the signal to improve the feature differentiation. Then, a dynamic window statistics method is introduced to construct comprehensive dynamic feature vectors to capture the transient changes in fault signals. Finally, the APABC is used to optimize the SVM classifier parameters to improve the classification performance and avoid the local optimum problem. The experimental results show that the method achieves an average accuracy of 99.091% in the complex fault diagnosis of switching power supplies, which is 21.8 percentage points higher than that of the traditional spectrum analysis method (77.273%). This study provides an efficient solution for the accurate diagnosis of complex fault modes in switching power supplies. Full article

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

► Show Figures

Figure 1

20 pages, 6160 KiB

Open AccessArticle

A Computational Approach to Increasing the Antenna System’s Sensitivity in a Doppler Radar Designed to Detect Human Vital Signs in the UHF-SHF Frequency Ranges

by David Vatamanu and Simona Miclaus

Sensors 2025, 25(10), 3235; https://doi.org/10.3390/s25103235 - 21 May 2025

Viewed by 378

Abstract

In the context of Doppler radar, studies have examined the changes in the phase shift of the S₂₁ transmission coefficient related to minute movements of the human chest as a response to breathing or heartbeat. Detecting human vital signs remains a challenge, [...] Read more.

In the context of Doppler radar, studies have examined the changes in the phase shift of the S₂₁ transmission coefficient related to minute movements of the human chest as a response to breathing or heartbeat. Detecting human vital signs remains a challenge, especially when obstacles interfere with the attempt to detect the presence of life. The sensitivity of a measurement system’s perception of vital signs is highly dependent on the monitoring systems and antennas that are used. The current work proposes a computational approach that aims to extract an empirical law of the dependence of the phase shift of the transmission coefficient (S₂₁) on the sensitivity at reception, based upon a set of four parameters. These variables are as follows: (a) the frequency of the continuous wave utilized; (b) the antenna type and its gain/directivity; (c) the electric field strength distribution on the chest surface (and its average value); and (d) the type of material (dielectric properties) impacted by the incident wave. The investigated frequency range is (1–20) GHz, while the simulations are generated using a doublet of dipole or gain-convenient identical Yagi antennas. The chest surface is represented by a planar rectangle that moves along a path of only 3 mm, with a step of 0.3 mm, mimicking respiration movement. The antenna–target system is modeled in the computational space in each new situation considered. The statistics illustrate the multiple regression function, empirically extracted. This enables the subsequent building of a continuous-wave bio-radar Doppler system with controlled and improved sensitivity. Full article

(This article belongs to the Special Issue Electromagnetic Waves, Antennas and Sensor Technologies in Modern Biomedical and Environmental Applications)

► Show Figures

Figure 1

20 pages, 8673 KiB

Open AccessCommunication

Harnessing Fast Fourier Transform for Rapid Community Travel Distance and Step Estimation in Children with Duchenne Muscular Dystrophy

by Erik K. Henricson and Albara Ah Ramli

Sensors 2025, 25(10), 3234; https://doi.org/10.3390/s25103234 - 21 May 2025

Viewed by 89

Abstract

Accurate estimation of gait characteristics, including step length, step velocity, and travel distance, is critical for assessing mobility in toddlers, children, and teens with Duchenne muscular dystrophy (DMD) and typically developing (TD) peers. This study introduces a novel method leveraging Fast Fourier Transform [...] Read more.

Accurate estimation of gait characteristics, including step length, step velocity, and travel distance, is critical for assessing mobility in toddlers, children, and teens with Duchenne muscular dystrophy (DMD) and typically developing (TD) peers. This study introduces a novel method leveraging Fast Fourier Transform (FFT)-derived step frequency from a single waist-worn consumer-grade accelerometer to predict gait parameters efficiently. The proposed FFT-based step frequency detection approach, combined with regression-derived stride length estimation, enables precise measurement of temporospatial gait features across various walking and running speeds. Our model, developed from a diverse cohort of children aged 3–16, demonstrated high accuracy in step length estimation (

R^{2} = 0.92

,

R M S E = 0.06

m) using only step frequency and height as inputs. Comparative analysis with ground-truth observations and AI-driven Walk4Me models validated the FFT-based method, showing strong agreement across step count, step frequency, step length, step velocity, and travel distance metrics. The results highlight the feasibility of using widely available mobile devices for gait assessment in real-world settings, offering a scalable solution for monitoring disease progression and mobility changes in individuals with DMD. Future work will focus on refining model performance and expanding applicability to additional movement disorders. Full article

(This article belongs to the Section Biomedical Sensors)

► Show Figures

Figure 1

22 pages, 6622 KiB

Open AccessArticle

Preliminary Development of a Database for Detecting Active Mounting Behaviors Using Signals Acquired from IoT Collars in Free-Grazing Cattle

by Miguel Guarda-Vera and Carlos Muñoz-Poblete

Sensors 2025, 25(10), 3233; https://doi.org/10.3390/s25103233 - 21 May 2025

Viewed by 96

Abstract

This study presents the development of a database for detecting active mounts, utilizing IoT collars equipped with Inertial Measurement Units (IMUs) installed on eight Holstein Friesian cows, along with video recordings from a long-range PTZ camera mounted in a central pole for event [...] Read more.

This study presents the development of a database for detecting active mounts, utilizing IoT collars equipped with Inertial Measurement Units (IMUs) installed on eight Holstein Friesian cows, along with video recordings from a long-range PTZ camera mounted in a central pole for event labeling in the natural environment when free grazing. The resulting database comprises 415 labeled events associated with various behaviors, containing acceleration signals in both the Body and World Frame of reference and gyroscope signals. A Support Vector Machine (SVM) algorithm is implemented to evaluate the effectiveness of the dataset in detecting active mounts and to compare training performance using both frames. The algorithm achieves an average F1 Score of 88.6% for the World Frame of reference, showing a significant improvement compared to the algorithm trained with Body Frame (78.6%) when both are trained with the same 112 features. After applying feature selection using Sequential Backward Selection (SBS), the SVM exhibits a minor performance difference between the F1 Score obtained with the two reference frames (89.7% World Frame vs. 91.5% Body Frame). This study provides a public dataset and a replicable methodology, facilitating future research on movement-based behavior classification in cattle. Full article

(This article belongs to the Special Issue Advances in Sensing-Based Animal Biomechanics)

► Show Figures

Figure 1

16 pages, 4869 KiB

Open AccessArticle

Cellulose Nanofibril-Based Triboelectric Nanogenerators Enhanced by Isoreticular Metal-Organic Frameworks for Long-Term Motion Monitoring

by Mingli Shang, Yan Zong and Xiujun Zhang

Sensors 2025, 25(10), 3232; https://doi.org/10.3390/s25103232 - 21 May 2025

Viewed by 115

Abstract

Cellulose nanofibril (CNF) is a sort of novel nanomaterial directly extracted from plant resources, inheriting the advantages of cellulose as a cheap, green and renewable material for the development of new-generation eco-friendly electronics. In recent years, CNF-based triboelectric nanogenerator (TENG) has attracted increasing [...] Read more.

Cellulose nanofibril (CNF) is a sort of novel nanomaterial directly extracted from plant resources, inheriting the advantages of cellulose as a cheap, green and renewable material for the development of new-generation eco-friendly electronics. In recent years, CNF-based triboelectric nanogenerator (TENG) has attracted increasing research interests, as the unique chemical, morphological, and electrical properties of CNF render the device with considerable flexibility, mechanical strength, and triboelectric output. In this study, we explore the use of isoreticular metal-organic frameworks (IRMOF) as functional filler to improve the performance of CNF based TENGs. Two types of IRMOFs that own the same network topology, namely IRMOF-1 and its aminated version IRMOF-3, are embedded with CNF to fabricated TENGs; their contribution to triboelectric output enhancement, including the roughness effect induced by large particles as well as the charge induction effect arisen from -NH₂ groups, are discussed. The performance-enhanced CNF-based TENG with 0.6 wt.% of IRMOF-3 is utilized to harvest mechanical energy from human activities and charge commercial capacitors, from which the electrical energy is sufficient to light up light-emitting diodes (LEDs) and drive low-power electronic devices. In addition, a locomotor analysis system is established by assembling the above TENGs and capacitors into a 3 × 3 sensing array, which allowed signal extraction from each sensing unit to display a motion distribution map. These results demonstrate the great potential of CNF/IRMOF-based TENGs for development of self-powered sensing devices for long-term motion monitoring. Full article

(This article belongs to the Section Electronic Sensors)

► Show Figures

Graphical abstract

25 pages, 6337 KiB

Open AccessArticle

Improved YOLOv8 Network of Aircraft Target Recognition Based on Synthetic Aperture Radar Imaging Feature

by Xing Wang, Wen Hong, Yunqing Liu, Guanyu Yan, Dongmei Hu and Qi Jing

Sensors 2025, 25(10), 3231; https://doi.org/10.3390/s25103231 - 21 May 2025

Viewed by 112

Abstract

The grayscale images of passenger aircraft targets obtained via Synthetic Aperture Radar (SAR) have problems such as complex airport backgrounds, significant speckle noise, and variable scales of targets. Most of the existing deep learning-based target recognition algorithms for SAR images are transferred from [...] Read more.

The grayscale images of passenger aircraft targets obtained via Synthetic Aperture Radar (SAR) have problems such as complex airport backgrounds, significant speckle noise, and variable scales of targets. Most of the existing deep learning-based target recognition algorithms for SAR images are transferred from optical images, and it is difficult for them to extract the multi-dimensional features of targets comprehensively. To overcome these challenges, we proposed three enhanced methods for interpreting aircraft targets based on YOLOv8. First, we employed the Shi–Tomasi corner detection algorithm and the Enhanced Lee filtering algorithm to convert grayscale images into RGB images, thereby improving detection accuracy and efficiency. Second, we augmented the YOLOv8 model with an additional detection branch, which includes a detection head featuring the Coordinate Attention (CA) mechanism. This enhancement boosts the model’s capability to detect small and multi-scale aircraft targets. Third, we integrated the Swin Transformer mechanism into the YOLOv8 backbone, forming the C2f-SWTran module that better captures long-range dependencies in the feature map. We applied these improvements to two datasets: the ISPRS-SAR-aircraft dataset and the SAR-Aircraft-1.0 dataset. The experimental results demonstrated that our methods increased the mean Average Precision (mAP_50~95) by 2.4% and 3.4% over the YOLOv8 baseline, showing competitive advantages over other deep learning-based object detection algorithms. Full article

(This article belongs to the Section Sensing and Imaging)

► Show Figures

Figure 1

8 pages, 188 KiB

Open AccessEditorial

Special Issue on Advanced Sensing and Control Technologies in Power Electronics

by Krzysztof Bernacki and Mateja Novak

Sensors 2025, 25(10), 3230; https://doi.org/10.3390/s25103230 - 21 May 2025

Viewed by 110

Abstract

Power electronics performs a key role in today’s energy systems, and its development is increasingly combined with advanced sensor-critical techniques and novel control algorithms that significantly enhance system adaptability, reliability, and robustness against disturbances and cyber threats [...] Full article

(This article belongs to the Special Issue Advanced Sensing and Control Technologies in Power Electronics)

15 pages, 2420 KiB

Open AccessArticle

Performance Comparison of Multipixel Biaxial Scanning Direct Time-of-Flight Light Detection and Ranging Systems With and Without Imaging Optics

by Konstantin Albert, Manuel Ligges, Andre Henschke, Jennifer Ruskowski, Menaka De Zoysa, Susumu Noda and Anton Grabmaier

Sensors 2025, 25(10), 3229; https://doi.org/10.3390/s25103229 - 21 May 2025

Viewed by 109

Abstract

The laser pulse detection probability of a scanning direct time-of-flight light detection and ranging (LiDAR) measurement is evaluated based on the optical signal distribution on a multipixel single photon avalanche diode (SPAD) array. These detectors intrinsically suffer from dead-times after the successful detection [...] Read more.

The laser pulse detection probability of a scanning direct time-of-flight light detection and ranging (LiDAR) measurement is evaluated based on the optical signal distribution on a multipixel single photon avalanche diode (SPAD) array. These detectors intrinsically suffer from dead-times after the successful detection of a single photon and, thus, allow only for limited counting statistics when multiple returning laser photons are imaged on a single pixel. By blurring the imaged laser spot, the transition from single-pixel statistics with high signal intensity to multipixel statistics with less signal intensity is examined. Specifically, a comparison is made between the boundary cases in which (i) the returning LiDAR signal is focused through optics onto a single pixel and (ii) the detection is performed without lenses using all available pixels on the sensor matrix. The omission of imaging optics reduces the overall system size and minimizes optical transfer losses, which is crucial given the limited laser emission power due to safety standards. The investigation relies on a photon rate model for interfering (background) and signal light, applied to a simulated first-photon sensor architecture. For single-shot scenarios that reflect the optimal use of the time budget in scanning LiDAR systems, the lens-less and blurred approaches can achieve comparable or even superior results to the focusing system. This highlights the potential of fully solid-state scanning LiDAR systems utilizing optical phase arrays or multidirectional laser chips. Full article

(This article belongs to the Special Issue SPAD-Based Sensors and Techniques for Enhanced Sensing Applications)

► Show Figures

Graphical abstract

14 pages, 2386 KiB

Open AccessArticle

An Ultra-Sensitive Colorimetric Sensing Platform for Simultaneous Detection of Moxifloxacin/Ciprofloxacin and Cr(III) Ions Based on Ammonium Thioglycolate Functionalized Gold Nanoparticles

by Lihua Zhang, Jiang Li, Juan Wang, Xu Yan, Jinping Song and Feng Feng

Sensors 2025, 25(10), 3228; https://doi.org/10.3390/s25103228 - 21 May 2025

Viewed by 99

Abstract

Water pollution by antibiotics and heavy metals threatens the ecological environment and human health globally, yet there is no rapid method to detect multiple antibiotics and metal ions simultaneously. A simple, fast, and ultra-sensitive colorimetric chemosensor for the simultaneous detection of moxifloxacin (MOX), [...] Read more.

Water pollution by antibiotics and heavy metals threatens the ecological environment and human health globally, yet there is no rapid method to detect multiple antibiotics and metal ions simultaneously. A simple, fast, and ultra-sensitive colorimetric chemosensor for the simultaneous detection of moxifloxacin (MOX), ciprofloxacin (CIP), and Cr(III) based on the aggregation of ammonium thioglycolate (ATG)-functionalized gold nanoparticles (ATG-AuNPs) was developed. Following the addition of MOX, CIP, and Cr(III), a color change in the solution was observed from wine-red to blue-grey. The UV–Vis signal of the ATG-AuNPs system blended with MOX, CIP, and Cr(III) in the range of 0~200 µM, 0~100 µM, and 0~5 µM was assessed and measured with detection limits (LODs) of 1.57 µM, 1.30 µM, and 57.1 nM calculated by 3σ/S, respectively. Therefore, this system has the potential to act as an effective colorimetric chemosensor for simultaneously detecting MOX, CIP, and Cr(III) in complex environmental systems. Full article

(This article belongs to the Section Nanosensors)

► Show Figures

Figure 1

20 pages, 4672 KiB

Open AccessArticle

Industrial-Grade Graphene Films as Distributed Temperature Sensors

by Francesco Siconolfi, Gabriele Cavaliere, Sarah Sibilia, Francesco Cristiano, Gaspare Giovinco and Antonio Maffucci

Sensors 2025, 25(10), 3227; https://doi.org/10.3390/s25103227 - 21 May 2025

Viewed by 68

Abstract

This paper investigates the feasibility of a multi-purpose use of thin films of industrial-grade graphene, adopted initially to realize advanced coatings for thermal management or electromagnetic shielding. Indeed, it is demonstrated that such coatings can be conveniently used as distributed temperature sensors based [...] Read more.

This paper investigates the feasibility of a multi-purpose use of thin films of industrial-grade graphene, adopted initially to realize advanced coatings for thermal management or electromagnetic shielding. Indeed, it is demonstrated that such coatings can be conveniently used as distributed temperature sensors based on the sensitivity of their electrical resistance to temperature. The study is carried out by characterizing three nanomaterials differing in the percentage of graphene nanoplatelets in the temperature range from −40 °C to +60 °C. The paper demonstrates the presence of a reproducible and linear negative temperature coefficient behavior, with a temperature coefficient of the resistance of the order of

- 1.5 \cdot 10^{- 3}

{° C}^{- 1}

. A linear sensor model is then developed and validated through an uncertainty-based approach, yielding a temperature prediction uncertainty of approximately ±2 °C. Finally, the robustness of the sensor concerning moderate environmental variations is verified, as the errors introduced by relative humidity values in the range from 40% to 60% are included in the model’s uncertainty bounds. These results suggest the realistic possibility of adding temperature-sensing capabilities to these graphene coatings with minimal increase in complexity and cost. Full article

(This article belongs to the Section Nanosensors)

► Show Figures

Figure 1

15 pages, 2549 KiB

Open AccessArticle

Automated Implementation of the Edinburgh Visual Gait Score (EVGS)

by Ishaasamyuktha Somasundaram, Albert Tu, Ramiro Olleac, Natalie Baddour and Edward D. Lemaire

Sensors 2025, 25(10), 3226; https://doi.org/10.3390/s25103226 - 21 May 2025

Viewed by 73

Abstract

The Edinburgh Visual Gait Score (EVGS) is a commonly used clinical scale for assessing gait abnormalities, providing insight into diagnosis and treatment planning. However, its manual implementation is resource-intensive and requires time, expertise, and a controlled environment for video recording and analysis. To [...] Read more.

The Edinburgh Visual Gait Score (EVGS) is a commonly used clinical scale for assessing gait abnormalities, providing insight into diagnosis and treatment planning. However, its manual implementation is resource-intensive and requires time, expertise, and a controlled environment for video recording and analysis. To address these issues, an automated approach for scoring the EVGS was developed. Unlike past methods dependent on controlled environments or simulated videos, the proposed approach integrates pose estimation with new algorithms to handle operational challenges present in the dataset, such as minor camera movement during sagittal recordings, slight zoom variations in coronal views, and partial visibility (e.g., missing head) in some videos. The system uses OpenPose for pose estimation and new algorithms for automatic gait event detection, stride segmentation, and computation of the 17 EVGS parameters across the sagittal and coronal planes. Evaluation of gait videos of patients with cerebral palsy showed high accuracy for parameters such as hip and knee flexion but a need for improvement in pelvic rotation and hindfoot alignment scoring. This automated EVGS approach can minimize the workload for clinicians through the introduction of automated, rapid gait analysis and enable mobile-based applications for clinical decision-making. Full article

(This article belongs to the Section Biomedical Sensors)

► Show Figures

Figure 1

23 pages, 3970 KiB

Open AccessArticle

Application of Neural Networks to Analyse the Spatial Distribution of Bicycle Traffic Before, During and After the Closure of the Mill Road Bridge in Cambridgeshire, United Kingdom

by Shohel Amin

Sensors 2025, 25(10), 3225; https://doi.org/10.3390/s25103225 - 20 May 2025

Viewed by 340

Abstract

Traffic congestions due to construction and maintenance works of road infrastructure cause travel delays, unpredictability and less tolerant road users. Bicyclists are more flexible with road closures, shifting to alternative routes, public transport and other active transport depending on the infrastructure, quality and [...] Read more.

Traffic congestions due to construction and maintenance works of road infrastructure cause travel delays, unpredictability and less tolerant road users. Bicyclists are more flexible with road closures, shifting to alternative routes, public transport and other active transport depending on the infrastructure, quality and transport services. However, the mixed traffic environment near road closures increases the safety risks for bicyclists. Traditional traffic monitoring systems rely on costly and demanding intrusive sensors. The application of wireless sensors and machine learning algorithms can enhance the analysis and prediction ability of traffic distribution and characteristics in the proximity of road closures. This paper applies artificial neural networks (ANNs) coupled with a Generalised Delta Rule (GDR) algorithm to analyse the sensor traffic data before, during and after the closure of the Mill Road Bridge in Cambridge City in the United Kingdom. The ANN models show that the traffic volume of motorbikes (44%) and buses (34%) and the proximity of Mill Road Bridge (39%) are significant factors affecting bicycle traffic before the closure. During the bridge closure, the proximity of the bridge (99%) and traffic volume of large rigid vehicles (51%) are the most important factors of bicycle distribution in nearby streets leading cyclists to unsafe detours. After the reopening of the Mill Road Bridge, unclear signage caused continued traffic impact, with motorbikes (17%) and large vehicles (24%) playing the most significant role in the spatial distribution of bicycle traffic. This paper emphasises safety concerns from mixed traffic and highlights the importance of cost-effective sensor-based traffic monitoring and analysis of the sensor data using neural networks. Full article

(This article belongs to the Section Physical Sensors)

► Show Figures

Figure 1

18 pages, 2298 KiB

Open AccessArticle

Driver Steering Intention Prediction for Human-Machine Shared Systems of Intelligent Vehicles Based on CNN-GRU Network

by Chen Zhou, Fan Zhang, Edric John Cruz Nacpil, Zheng Wang and Fei-Xiang Xu

Sensors 2025, 25(10), 3224; https://doi.org/10.3390/s25103224 - 20 May 2025

Viewed by 326

Abstract

In order to mitigate human-machine conflicts and optimize shared control strategy in advance, it is essential for the shared control system to understand and predict driver behavior. This paper proposes a method for predicting driver steering intention with a CNN-GRU hybrid machine learning [...] Read more.

In order to mitigate human-machine conflicts and optimize shared control strategy in advance, it is essential for the shared control system to understand and predict driver behavior. This paper proposes a method for predicting driver steering intention with a CNN-GRU hybrid machine learning model. The convolutional neural network (CNN) layer extracts features from the stochastic driver behavior, which is input to the gated-recurrent-unit (GRU) layer. And the driver’s steering intention is forecasted based on the GRU model. Our study was conducted using a driving simulator to observe the lateral control behaviors of 18 participants in four different driving circumstances. Finally, the efficiency of the suggested prediction approach was evaluated employing long-short-term-memory, GRU, CNN, Transformer, and back propagation networks. Experimental results demonstrated that the proposed CNN-GRU model performs significantly better than baseline models. Compared with the GRU network, the CNN-GRU network reduced the RMSE, MAE, and MAPE of the driver’s input torque by 33.22%, 32.33%, and 35.86%, respectively. The proposed prediction method also possesses adaptability to different driver behaviors. Full article

(This article belongs to the Special Issue Connected and Autonomous Vehicles: Trends, Applications and Security Challenges)

► Show Figures

Figure 1

28 pages, 6459 KiB

Open AccessArticle

Soil Porosity Detection Method Based on Ultrasound and Multi-Scale Feature Extraction

by Hang Xing, Zeyang Zhong, Wenhao Zhang, Yu Jiang, Xinyu Jiang, Xiuli Yang, Weizi Cai, Shuanglong Wu and Long Qi

Sensors 2025, 25(10), 3223; https://doi.org/10.3390/s25103223 - 20 May 2025

Viewed by 144

Abstract

Soil porosity, as an essential indicator for assessing soil quality, plays a key role in guiding agricultural production, so it is beneficial to detect soil porosity. However, the currently available methods do not apply to high-precision and rapid detection of soil with a [...] Read more.

Soil porosity, as an essential indicator for assessing soil quality, plays a key role in guiding agricultural production, so it is beneficial to detect soil porosity. However, the currently available methods do not apply to high-precision and rapid detection of soil with a black-box nature in the field, so this paper proposes a soil porosity detection method based on ultrasound and multi-scale CNN-LSTM. Firstly, a series of ring cutter soil samples with different porosities were prepared manually to simulate soil collected in the field using a ring cutter, followed by ultrasonic signal acquisition of the soil samples. The acquired signals were subjected to three kinds of data augmentation processes to enrich the dataset: adding Gaussian white noise, time shift transformation, and random perturbation. Since the collected ultrasonic signals belong to long-time series data and there are different frequency and sequence features, this study constructs a multi-scale CNN-LSTM deep neural network model using large convolution kernels based on the idea of multi-scale feature extraction, which uses multiple large convolution kernels of different sizes to downsize the collected ultra-long time series data and extract local features in the sequences, and combining the ability of LSTM to capture global and long-term dependent features enhances the feature expression ability of the model. The multi-head self-attention mechanism is added at the end of the model to infer the before-and-after relationship of the sequence data to improve the degradation of the model performance caused by waveform distortion. Finally, the model was trained, validated, and tested using ultrasonic signal data collected from soil samples to demonstrate the accuracy of the detection method. The model has a coefficient of determination of 0.9990 for detecting soil porosity, with a percentage root mean square error of only 0.66%. It outperforms other advanced comparative models, making it very promising for application. Full article

(This article belongs to the Section Smart Agriculture)

► Show Figures

Figure 1

21 pages, 2069 KiB

Open AccessArticle

Assessing Physiological Stress Responses in Student Nurses Using Mixed Reality Training

by Kamelia Sepanloo, Daniel Shevelev, Young-Jun Son, Shravan Aras and Janine E. Hinton

Sensors 2025, 25(10), 3222; https://doi.org/10.3390/s25103222 - 20 May 2025

Viewed by 143

Abstract

This study explores nursing students’ stress responses while they are being trained in a mixed reality (MR) setting that replicates highly stressful clinical scenarios. Using measurements of physiological indices such as heart rate, electrodermal activity, and skin temperature, the study assesses the level [...] Read more.

This study explores nursing students’ stress responses while they are being trained in a mixed reality (MR) setting that replicates highly stressful clinical scenarios. Using measurements of physiological indices such as heart rate, electrodermal activity, and skin temperature, the study assesses the level of stress when the students interact with digital patients whose vital signs and symptoms interact dynamically to respond to student inputs. The simulation consists of six segments, during which critical events like hypotension and hypoxia occur, and the patient’s condition changes based on the nurse’s clinical decisions. Machine learning algorithms were then used to analyze the nurse’s physiological data and to classify different levels of stress. Among the models tested, the Stacking Classifier demonstrated the highest classification accuracy of 96.4%, outperforming both Random Forest (96.18%) and Gradient Boosting (95.35%). The results showed clear patterns of stress during the simulation segments. Statistical analysis also found significant differences in stress responses and identified key physiological markers linked to each stress level. This pioneering study demonstrates the effectiveness of MR as a training tool for healthcare professionals in high-pressured scenarios and lays the groundwork for further studies on stress management, adaptive training procedures, and real-time detection and intervention in MR-based nursing training. Full article

(This article belongs to the Special Issue Sensors for Physiological Monitoring and Digital Health)

► Show Figures

Figure 1

18 pages, 4018 KiB

Open AccessArticle

Super-Resolution Image Optimisation Based on Gradient Iterative Fast Diffraction-Free Spot Algorithm

by Chen Yu, Ying Liu, Linhan Li, Guangpeng Zhou, Boshi Dang, Jie Du, Junlin Ma and Site Zhang

Sensors 2025, 25(10), 3221; https://doi.org/10.3390/s25103221 - 20 May 2025

Viewed by 98

Abstract

Diffraction significantly deteriorates the quality of the laser image, causing severe degradation that undermines the theoretical performance parameters of the autofocus system. In this paper, we conduct a comprehensive analysis of the non-uniform features of the images. To enhance the imaging quality of [...] Read more.

Diffraction significantly deteriorates the quality of the laser image, causing severe degradation that undermines the theoretical performance parameters of the autofocus system. In this paper, we conduct a comprehensive analysis of the non-uniform features of the images. To enhance the imaging quality of each individual image, we propose a de-diffraction algorithm based on gradient iteration. This algorithm is capable of rapidly removing the interference spots resulting from diffraction and restoring the distorted laser spots. By doing so, it effectively eliminates the inevitable reduction in the autofocus resolution and focusing accuracy caused by diffraction. Furthermore, the proposed calculation model for the intra-localisation interval significantly improves the convergence of the iterative calculation process. Through experiments, it has been verified that, under the same conditions, the interlayer resolution between the reflective surfaces of the samples processed using this algorithm is increased to a quarter of the original value. This remarkable improvement in resolution, which far exceeds the microscope’s inherent resolution, demonstrates that the algorithm successfully achieves super-resolution for the microscope. Full article

(This article belongs to the Section Optical Sensors)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Sensors, Volume 25, Issue 10 (May-2 2025) – 295 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI