Search

In the production management of agriculture, accurate fruit counting plays a vital role in the orchard yield estimation and appropriate production decisions. Although recent tracking-by-detection algorithms have emerged as a promising fruit-counting method, they still cannot completely avoid fruit occlusion and light variations in complex orchard environments, and it is difficult to realize automatic and accurate apple counting. In this paper, a video-based multiple-object tracking method, MR-SORT (Multiple Rematching SORT), is proposed based on the improved YOLOv8 and BoT-SORT. First, we propose the AD-YOLO model, which aims to reduce the number of incorrect detections during object tracking. In the YOLOv8s backbone network, an Omni-dimensional Dynamic Convolution (ODConv) module is used to extract local feature information and enhance the model's ability better; a Global Attention Mechanism (GAM) is introduced to improve the detection ability of a foreground object (apple) in the whole image; a Soft Spatial Pyramid Pooling Layer (SSPPL) is designed to reduce the feature information dispersion and increase the sensory field of the network. Then, the improved BoT-SORT algorithm is proposed by fusing the verification mechanism, SURF feature descriptors, and the Vector of Local Aggregate Descriptors (VLAD) algorithm, which can match apples more accurately in adjacent video frames and reduce the probability of ID switching in the tracking process. The results show that the mAP metrics of the proposed AD-YOLO model are 3.1% higher than those of the YOLOv8 model, reaching 96.4%. The improved tracking algorithm has 297 fewer ID switches, which is 35.6% less than the original algorithm. The multiple-object tracking accuracy of the improved algorithm reached 85.6%, and the average counting error was reduced to 0.07. The coefficient of determination R2 between the ground truth and the predicted value reached 0.98. The above metrics show that our method can give more accurate counting results for apples and even other types of fruit.

Current super-resolution algorithms exhibit limitations when processing noisy remote sensing images rich in surface information, as they tend to amplify noise during the recovery of high-frequency signals. To mitigate this issue, this paper presents a novel approach that incorporates the concept of compressed sensing and explores the super-resolution problem of remote sensing images for space cameras, particularly for high-speed imaging systems. The proposed algorithm employs K-singular value decomposition (K-SVD) to jointly train high- and low-resolution image blocks, updating them column by column to obtain overcomplete dictionary pairs. This approach compensates for the deficiency of fixed dictionaries in the original algorithm. In the process of dictionary updating, we innovatively integrate the circle chaotic mapping into the solution process of the dictionary sequence, replacing pseudorandom numbers. This integration facilitates balanced traversal and simplifies the search for global optimal solutions. For the optimization problem of sparse coefficients, we utilize the orthogonal matching pursuit method (OMP) instead of the L1 norm convex optimization method used in most reconstruction techniques, thereby complementing the K-SVD dictionary update algorithm. After upscaling and denoising the image using the dictionary pair mapping relationship, we further emphasize image edge details with local gradients as constraints. When compared with various representative super-resolution algorithms, our algorithm effectively filters out noise and stains in low-resolution images. It not only performs well visually but also stands out in objective evaluation indicators such as the peak signal-to-noise ratio and information entropy. The experimental results validate the effectiveness of the proposed method in super-resolution remote sensing images, yielding high-quality remote sensing image data.

Line charts are prevalent in scientific documents and commercial data visualization, serving as essential tools for conveying data trends. Automatic detection and tracing of line paths in these charts is crucial for downstream tasks such as data extraction, chart quality assessment, plagiarism detection, and visual question answering. However, line graphs present unique challenges due to their complex backgrounds and diverse curve styles, including solid, dashed, and dotted lines. Existing curve detection algorithms struggle to address these challenges effectively. In this paper, we propose ChartLine, a novel network designed for detecting and tracing curves in line graphs. Our approach integrates a Spatial-Sequence Attention Feature Pyramid Network (SSA-FPN) in both the encoder and decoder to capture rich hierarchical representations of curve structures and boundary features. The model incorporates a Spatial-Sequence Fusion (SSF) module and a Channel Multi-Head Attention (CMA) module to enhance intra-class consistency and inter-class distinction. We evaluate ChartLine on four line chart datasets and compare its performance against state-of-the-art curve detection, edge detection, and semantic segmentation methods. Extensive experiments demonstrate that our method significantly outperforms existing algorithms, achieving an F-measure of 94% on a synthetic dataset.

Point cloud semantic segmentation is crucial for identifying and analyzing transmission lines. Due to the number of point clouds being huge, complex scenes, and unbalanced sample proportion, the mainstream machine learning methods of point cloud segmentation cannot provide high efficiency and accuracy when extending to transmission line scenes. This paper proposes a filter-assisted airborne point cloud semantic segmentation for transmission lines. First, a large number of ground point clouds is identified by introducing the well-developed cloth simulation filter to alleviate the impact of the imbalance of the target object proportion on the classifier's performance. The multi-dimensional features are then defined, and the classification model is trained to achieve the multi-element semantic segmentation of the transmission line scene. The experimental results and analysis indicate that the proposed filter-assisted algorithm can significantly improve the semantic segmentation performance of the transmission line point cloud, enhancing both the point cloud segmentation efficiency and accuracy by more than 25.46% and 3.15%, respectively. The filter-assisted point cloud semantic segmentation method reduces the volume of sample data, the number of sample classes, and the sample imbalance index in power line scenarios to a certain extent, thereby improving the classification accuracy of classifiers and reducing time consumption. This research holds significant theoretical reference value and engineering application potential for scene reconstruction and intelligent understanding of airborne laser point cloud transmission lines.

To better cope with stress in emergencies, emergency personnel undergo virtual reality (VR) stress training. Such training typically includes visual, auditory and sometimes tactile impressions, whereas olfactory stimuli are mostly neglected. This concept paper therefore examines whether odors might be beneficial for further enhancing the experience of presence and immersion into a simulated environment. The aim is to demonstrate the benefits of VR civilian stress training for emergency personnel and to investigate the role of odors as stressors by manipulating the degree of perceived psychophysiological stress via olfactory impressions. Moreover, the current paper presents the development and validation of a convenient and portable fragrance dosing system that allows personalized odor presentation in VR. The presented system can transport reproducible small quantities of an air-fragrance mixture close to the human nose using piezoelectric stainless steel micropumps. The results of the fluidic system validation indicate that the micropump is suitable for releasing odors close to the nose with constant amounts of odor presentation. Furthermore, the theoretical background and the planned experimental design of VR stress training, including odor presentation via olfactory VR technology, are elucidated.

In the era of digital music proliferation, music genre classification has become a crucial task in music information retrieval. This paper proposes a novel channel-aware convolutional neural network (ECAS-CNN) designed to enhance the efficiency and accuracy of music genre recognition. By integrating an adaptive channel attention mechanism (ECA module) within the convolutional layers, the network significantly improves the extraction of key musical features. Extensive experiments were conducted on the GTZAN dataset, comparing the proposed ECAS-CNN with traditional convolutional neural networks. The results demonstrate that ECAS-CNN outperforms conventional methods across various performance metrics, including accuracy, precision, recall, and F1-score, particularly in handling complex musical features. This study validates the potential of ECAS-CNN in the domain of music genre classification and offers new insights for future research and applications.

The global increase in the population of Visually Impaired People (VIPs) underscores the rapidly growing demand for a robust navigation system to provide safe navigation in diverse environments. State-of-the-art VIP navigation systems cannot achieve the required performance (accuracy, integrity, availability, and integrity) because of insufficient positioning capabilities and unreliable investigations of transition areas and complex environments (indoor, outdoor, and urban). The primary reason for these challenges lies in the segregation of Visual Impairment (VI) research within medical and engineering disciplines, impeding technology developers' access to comprehensive user requirements. To bridge this gap, this paper conducts a comprehensive review covering global classifications of VI, international and regional standards for VIP navigation, fundamental VIP requirements, experimentation on VIP behavior, an evaluation of state-of-the-art positioning systems for VIP navigation and wayfinding, and ways to overcome difficulties during exceptional times such as COVID-19. This review identifies current research gaps, offering insights into areas requiring advancements. Future work and recommendations are presented to enhance VIP mobility, enable daily activities, and promote societal integration. This paper addresses the urgent need for high-performance navigation systems for the growing population of VIPs, highlighting the limitations of current technologies in complex environments. Through a comprehensive review of VI classifications, VIPs' navigation standards, user requirements, and positioning systems, this paper identifies research gaps and offers recommendations to improve VIP mobility and societal integration.

Battery-powered sensor nodes encounter substantial energy constraints, especially in linear wireless sensor network (LWSN) applications like border surveillance and road, bridge, railway, powerline, and pipeline monitoring, where inaccessible locations exacerbate battery replacement challenges. Addressing these issues is crucial for extending a network's lifetime and reducing operational costs. This paper presents a comprehensive analysis of the factors affecting WSN energy consumption at the node and network levels, alongside effective energy management strategies for prolonging the WSN's lifetime. By categorizing existing strategies into node energy reduction, network energy balancing, and energy replenishment, this study assesses their effectiveness when implemented in LWSN applications, providing valuable insights to assist engineers during the design of green and energy-efficient LWSN monitoring systems.

In the past few decades, there has been a notable technological advancement in geophysical sensors. In the case of magnetometry, several sensors were used, having the common feature of being miniaturized and lightweight, thus idoneous to be carried by UAVs in drone-borne magnetometric surveys. A common feature is that their sensitivity ranges from 0.1 to about 200 nT, thus not comparable to that of optically pumped, standard fluxgate or even proton magnetometers. However, their low cost, volume and weight remain very interesting features of these sensors. In fact, such sensors have the common feature of being very inexpensive, so new ways of making surveys using many of these sensors could be devised, in addition to the possibility, even with limited resources, of creating gradiometers by combining two or more of them. In this paper, we explore the range of applicability of small tri-axial magnetometers commonly used for attitude determination in several devices. We compare the results of surveys performed with standard professional geophysical instruments with those obtained using these sensors and find that in the presence of strongly magnetized sources, they succeeded in identifying the main anomalies.

Breast cancer is the most prevalent cancer among women globally, making early and accurate detection essential for effective treatment and improved survival rates. This paper presents a method designed to detect and localize breast cancer using deep learning, specifically convolutional neural networks. The approach classifies histological images of breast tissue as either tumor-positive or tumor-negative. We utilize several deep learning models, including a custom-built CNN, EfficientNet, ResNet50, VGG-16, VGG-19, and MobileNet. Fine-tuning was also applied to VGG-16, VGG-19, and MobileNet to enhance performance. Additionally, we introduce a novel deep learning model called MR_Net, aimed at providing a more accurate network for breast cancer detection and localization, potentially assisting clinicians in making informed decisions. This model could also accelerate the diagnostic process, enabling early detection of the disease. Furthermore, we propose a method for explainable predictions by generating heatmaps that highlight the regions within tissue images that the model focuses on when predicting a label, revealing the detection of benign, atypical, and malignant tumors. We evaluate both the quantitative and qualitative performance of MR_Net and the other models, also presenting explainable results that allow visualization of the tissue areas identified by the model as relevant to the presence of breast cancer.

This paper proposes a novel decentralized federated reinforcement learning (DFRL) framework that integrates deep reinforcement learning (DRL) with decentralized federated learning (DFL). The DFRL framework boosts efficient virtual instance scaling in Mobile Edge Computing (MEC) environments for 5G core network automation. It enables multiple MECs to collaboratively optimize resource allocation without centralized data sharing. In this framework, DRL agents in each MEC make local scaling decisions and exchange model parameters with other MECs, rather than sharing raw data. To enhance robustness against malicious server attacks, we employ a committee mechanism that monitors the DFL process and ensures reliable aggregation of local gradients. Extensive simulations were conducted to evaluate the proposed framework, demonstrating its ability to maintain cost-effective resource usage while significantly reducing blocking rates across diverse traffic conditions. Furthermore, the framework demonstrated strong resilience against adversarial MEC nodes, ensuring reliable operation and efficient resource management. These results validate the framework's effectiveness in adaptive and efficient resource management, particularly in dynamic and varied network scenarios.

This paper investigates methods that leverage physical contact between a robot's structure and its environment to enhance task performance, with a primary emphasis on improving precision. Two main approaches are examined: solving the inverse kinematics problem and employing quadratic programming, which offers computational efficiency by utilizing forward kinematics. Additionally, geometrical methods are explored to simplify robot assembly and reduce the complexity of control calculations. These approaches are implemented on a physical robotic platform and evaluated in real-time applications to assess their effectiveness. Through experimental evaluation, this study aims to understand how environmental contact can be utilized to enhance performance across various conditions, offering valuable insights for practical applications in robotics.

Due to the complex intersection and control of multiple structural systems, the hydrogeological conditions of the Laiyuan Basin in China are complex. The depth of research on the relationship between geological structure and groundwater migration needs to be improved. The supply relationship of each aquifer is still uncertain. This paper systematically conducts research on the characteristics of hydrogen and oxygen isotopes, and combines magnetotelluric impedance tensor decomposition and two-dimensional fine inversion technology to carry out fine exploration of the strata and structures in the Laiyuan Basin, as well as comprehensive characteristics of groundwater migration and replenishment. The results indicate the following: (i) The hydrogen and oxygen values all fall near the local meteoric water line, indicating that precipitation is the main groundwater recharge source. (ii) The excess deuterium decreased gradually from karst mountain to basin, and karst water and pore water experienced different flow processes. (iii) The structure characteristics of three main runoff channels are described by MT fine processing and inversion techniques. Finally, it is concluded that limestone water moved from the recharge to the discharge area, mixed with the deep dolomite water along the fault under the control of fault F2, and eventually rose to the surface of the unconsolidated sediment blocked by fault F1 to emerge into an ascending spring.

Underwater simultaneous localization and mapping (SLAM) is essential for effectively navigating and mapping underwater environments; however, traditional SLAM systems have limitations due to restricted vision and the constantly changing conditions of the underwater environment. This study thoroughly examined the underwater SLAM technology, particularly emphasizing the incorporation of deep learning methods to improve performance. We analyzed the advancements made in underwater SLAM algorithms. We explored the principles behind SLAM and deep learning techniques, examining how these methods tackle the specific difficulties encountered in underwater environments. The main contributions of this work are a thorough assessment of the research into the use of deep learning in underwater image processing and perception and a comparison study of standard and deep learning-based SLAM systems. This paper emphasizes specific deep learning techniques, including generative adversarial networks (GANs), convolutional neural networks (CNNs), long short-term memory (LSTM) networks, and other advanced methods to enhance feature extraction, data fusion, scene understanding, etc. This study highlights the potential of deep learning in overcoming the constraints of traditional underwater SLAM methods, providing fresh opportunities for exploration and industrial use.

The research on how to effectively extract the instantaneous characteristic components of non-stationary signals continues to be both a research hotspot and a very challenging topic. In this paper, a new method of multi-component decomposition is proposed to decompose a signal into finite mono-component signals and extract their Instantaneous Amplitude (IA), Instantaneous Phase (IP), and Instantaneous Frequency (IF), which is called Sinusoidal Fitting Decomposition (SFD). The proposed method can ensure that the IA extracted from the given signal must be positive, the IP is monotonically increasing, and the signal synthesized by both IA and IP must be mono-componential and smooth. It transforms the decomposition process into a synthesis iterative process and does not rely on any dictionary or basis function space or carry out the sifting operation. In addition, the proposed method can describe the instantaneous-frequency-amplitude characteristics of the signal very well on the time-frequency plane. The results of numerical simulation and the qualitative analysis of the amount of calculation show that the proposed method is effective.

Impact loads affect the operational performance and safety life of rolling equipment's connecting-shaft rotor system, even causing faults and accidents. Therefore, recognizing and investigating impact loads is of great significance. Hence, a load recognition method based on motor current information is proposed in this paper to recognize impact loads on the connecting-shaft rotor system. First, the fast Fourier transform is used to obtain the frequency domain information for the motor's current response signal from the rotor system load recognition test. Consequently, the required load response information can be presented more clearly using the singular value decomposition method to remove the power frequency components in the current signal. Then, wavelet packet decomposition is performed on the signal to generate energy analysis feature vectors. A qualitative recognition of the impact load on the system is achieved by learning vector quantization neural networks; the resulting load recognition results are good. These findings indicate that using the motor current as the analysis signal can solve the problem of the difficult layout for traditional vibration sensors in rolling sites. The preprocessing and recognition method of the current response signal can recognize the impact load, confirming the applicability and feasibility of the proposed method.

In-flight alignment is a critical milestone for inertial navigation system/global navigation satellite system (INS/GNSS) applications in unmanned aerial vehicles (UAVs). The traditional position integration formula for in-flight coarse alignment requires the GNSS velocity data to be valid throughout the alignment period, which greatly limits the engineering applicability of the method. In this paper, a new robust position integration optimization-based alignment (OBA) method for in-flight coarse alignment is presented to solve the problem of in-flight alignment under a prolonged ineffective GNSS. In this methodology, to achieve a higher alignment accuracy in case the GNSS is not effective throughout the alignment period, the integration of GNSS velocity into the local-level navigation frame is replaced by the GNSS position in the Earth-centered, Earth-fixed frame, which avoids the need for complete GNSS velocity data. The simulation and flight test results show that the new robust position integration method proposed in this paper achieves higher stability and robustness than the conventional position integration OBA method and can achieve an alignment accuracy of 0.2° even when the GNSS is partially time-invalidated. Thus, this greatly extends the application of the OBA method for in-flight alignment.

Effective detection techniques are important for wildlife monitoring and conservation applications and are especially helpful for species that live in complex environments, such as arboreal animals like koalas ( Phascolarctos cinereus ). The implementation of infrared cameras and drones has demonstrated encouraging outcomes, regardless of whether the detection was performed by human observers or automated algorithms. In the case of koala detection in eucalyptus plantations, there is a risk to spotters during forestry operations. In addition, fatigue and tedium associated with the difficult and repetitive task of checking every tree means automated detection options are particularly desirable. However, obtaining high detection rates with minimal false alarms remains a challenging task, particularly when there is low contrast between the animals and their surroundings. Koalas are also small and often partially or fully occluded by canopy, tree stems, or branches, or the background is highly complex. Biologically inspired vision systems are known for their superior ability in suppressing clutter and enhancing the contrast of dim objects of interest against their surroundings. This paper introduces a biologically inspired detection algorithm to locate koalas in eucalyptus plantations and evaluates its performance against ten other detection techniques, including both image processing and neural-network-based approaches. The nature of koala occlusion by canopy cover in these plantations was also examined using a combination of simulated and real data. The results show that the biologically inspired approach significantly outperformed the competing neural-network- and computer-vision-based approaches by over 27%. The analysis of simulated and real data shows that koala occlusion by tree stems and canopy can have a significant impact on the potential detection of koalas, with koalas being fully occluded in up to 40% of images in which koalas were known to be present. Our analysis shows the koala's heat signature is more likely to be occluded when it is close to the centre of the image (i.e., it is directly under a drone) and less likely to be occluded off the zenith. This has implications for flight considerations. This paper also describes a new accurate ground-truth dataset of aerial high-dynamic-range infrared imagery containing instances of koala heat signatures. This dataset is made publicly available to support the research community.

An accurate panorama depth estimation result is crucial to risk perception in autonomous driving practice. In this paper, an innovative framework is presented to address the challenges of imperfect observation and projection fusion in panorama depth estimation, enabling the accurate capture of distances from surrounding images in driving scenarios. First, the Patch Filling method is proposed to alleviate the imperfect observation of panoramic depth in autonomous driving scenarios, which constructs a panoramic depth map based on the sparse distance data provided by the 3D point cloud. Then, in order to tackle the distortion challenge faced by outdoor panoramic images, a method for image context learning, ViT-Fuse, is proposed and specifically designed for equirectangular panoramic views. The experimental results show that the proposed ViT-Fuse reduces the estimation error by 9.15% on average in driving scenarios compared with the basic method and exhibits more robust and smoother results on the edge details of the depth estimation maps.

This paper presents a comprehensive review of cardiorespiratory auscultation sensing devices (i.e., stethoscopes), which is useful for understanding the theoretical aspects and practical design notes. In this paper, we first introduce the acoustic properties of the heart and lungs, as well as a brief history of stethoscope evolution. Then, we discuss the basic concept of electret condenser microphones (ECMs) and a stethoscope based on them. Then, we discuss the microelectromechanical systems (MEMSs) technology, particularly focusing on piezoelectric transducer sensors. This paper comprehensively reviews sensing technologies for cardiorespiratory auscultation, emphasizing MEMS-based wearable designs in the past decade. To our knowledge, this is the first paper to summarize ECM and MEMS applications for heart and lung sound analysis.

This study examines the measurement of film thickness, curvature, and defects on the surface or inside of an optical element using a highly accurate and efficient method. This is essential to ensure their quality and performance. Existing methods are unable to simultaneously extract the three types of information: thickness, curvature, and defects. Spectral-domain optical coherence tomography (SD-OCT), a non-invasive imaging technique with imaging depths down to the millimeter scale, provides the possibility of detecting the optical element components' parameters. In this paper, we propose an error correction model for compensating delay differences in A-scan, field curvature, and aberration to improve the accuracy of system fitting measurements using SD-OCT. During data processing, we use the histogram-equalized gray stretching (IAH-GS) method to deal with strong reflections in the thin film layers inside the optics using individual A-scan averages. In addition, we propose a window threshold cutoff algorithm to accurately identify defects and boundaries in OCT images. Finally, the system is capable of rapidly detecting the thickness and curvature of film layers in optical elements with a maximum measurement depth of 4.508 mm, a diameter of 15 × 15 mm, a resolution of 5.69 microns, and a sampling rate of 70 kHz. Measurements were performed on different standard optical elements to verify the accuracy and reliability of the proposed method. To the best of our knowledge, this is the first time that thickness, curvature, and defects of an optical film have been measured simultaneously, with a thickness measurement accuracy of 1.924 µm, and with a difference between the calibrated and nominal curvature measurements consistently within 1%. We believe that this research will greatly advance the use of OCT technology in the testing of optical thin films, thereby improving productivity and product quality.

Underground mines are considered one of the riskiest facilities for human activities due to numerous accidents and geotechnical failures recorded worldwide over the last century, which have resulted in unsafe labor conditions, poor health outcomes, injuries, and fatalities. One significant cause of these accidents is the inadequate or nonexistent capacity for the real-time monitoring of safety conditions in underground mines. In this context, new emerging technologies linked to the Industry 4.0 paradigm, such as sensors, the Internet of Things (IoT), and LoRaWAN (Long Range Wide Area Network) wireless connectivity, are being implemented for planning the efficient, safe, and sustainable performance of underground mine labor environments. This paper studies the implementation of an ecosystem composed of IoT sensors and LoRa wireless connectivity in a data-acquisition system, which eliminates the need for expensive cabling and manual monitoring in mining operations. Laying cables in an underground mine necessitates cable support and protection against issues, such as machinery operations, vehicle movements, mine operator activities, and groundwater intrusion. As the underground mine expands, additional sensors typically require costly cable installations unless wireless connectivity is employed. The results of this review indicate that an IoT LoRaWAN-based wireless sensor network (WSN) provides real-time data under complex conditions, effectively transmitting data through physical barriers. This network presents an attractive low-cost solution with reliable, simple, scalable, secure, and competitive characteristics compared to cable installations and manually collected readings, which are more sporadic and prone to human error. Reliable data on the behavior of the underground mine enhances productivity by improving key performance indicators (KPIs), minimizing accident risks, and promoting sustainable environmental conditions for mine operators. Finally, the adoption of IoT sensors and LoRaWAN wireless connectivity technologies provides information of the underground mine in real-time, which supports better decisions by the mining industry managers, by ensuring compliance with safety regulations, improving the productive performance, and fostering a roadmap towards more environmentally friendly labor conditions.

The Low-Energy Adaptive Clustering Hierarchy (LEACH) protocol is a widely used method for managing energy consumption in Wireless Sensor Networks (WSNs). However, it has limitations that affect network longevity and performance. This paper presents an improved version of the LEACH protocol, termed MFG-LEACH, which incorporates the Mean Field Game (MFG) theory to optimize energy efficiency and network lifetime. The proposed MFG-LEACH protocol addresses the imbalances in energy consumption by modeling the interactions among nodes as a game, where each node optimizes its transmission energy based on the collective state of the network. We conducted extensive simulations to compare MFG-LEACH with Enhanced Zonal Stable Election Protocol (EZ-SEP), Energy-Aware Multi-Hop Routing (EAMR), and Balanced Residual Energy routing (BRE) protocols. The results demonstrate that MFG-LEACH significantly reduces energy consumption and increases the number of packets received across different node densities, thereby validating the effectiveness of our approach.

A novel detection model, MS-YOLO, is developed in this paper to improve the efficiency of drowning rescue operations. The model is lightweight, high in precision, and applicable for intelligent hardware platforms. Firstly, the MD-C2F structure is built to capture the subtle movements and posture changes in various aquatic environments, with a light weight achieved by introducing dynamic convolution (DcConv). To make the model perform better in small object detection, the EMA mechanism is incorporated into the MD-C2F. Secondly, the MSI-SPPF module is constructed to improve the performance in identifying the features of different scales and the understanding of complex backgrounds. Finally, the ConCat single-channel fusion is replaced by BiFPN weighted channel fusion to retain more feature information and remove the irrelevant information in drowning features. Relative to the Faster R-CNN, SSD, YOLOv6, YOLOv9, and YOLOv10, the MS-YOLO achieves an average accuracy of 86.4% in detection on a self-built dataset at an ultra-low computational cost of 7.3 GFLOPs.

This study addresses the challenges of human-robot interactions in real-time environments with adaptive field-programmable gate array (FPGA)-based accelerators. Predicting human posture in indoor environments in confined areas is a significant challenge for service robots. The proposed approach works on two levels: the estimation of human location and the robot's intention to serve based on the human's location at static and adaptive positions. This paper presents three methodologies to address these challenges: binary classification to analyze static and adaptive postures for human localization in indoor environments using the sensor fusion method, adaptive Simultaneous Localization and Mapping (SLAM) for the robot to deliver the task, and human-robot implicit communication. VLSI hardware schemes are developed for the proposed method. Initially, the control unit processes real-time sensor data through PIR sensors and multiple ultrasonic sensors to analyze the human posture. Subsequently, static and adaptive human posture data are communicated to the robot via Wi-Fi. Finally, the robot performs services for humans using an adaptive SLAM-based triangulation navigation method. The experimental validation was conducted in a hospital environment. The proposed algorithms were coded in Verilog HDL, simulated, and synthesized using VIVADO 2017.3. A Zed-board-based FPGA Xilinx board was used for experimental validation.

This study developed an automated language learning teaching assessment system based on electroencephalography (EEG) and differential language large models (LLMs), aimed at enhancing language instruction effectiveness by monitoring learners' cognitive states in real time and personalizing teaching content accordingly. Through detailed experimental design, the paper validated the system's application in various teaching tasks. The results indicate that the system exhibited high precision, recall, and accuracy in teaching effectiveness tests. Specifically, the method integrating differential LLMs with the EEG fusion module achieved a precision of 0.96, recall of 0.95, accuracy of 0.96, and an F1-score of 0.95, outperforming other automated teaching models. Additionally, ablation experiments further confirmed the critical role of the EEG fusion module in enhancing teaching quality and accuracy, providing valuable data support and theoretical basis for future improvements in teaching methods and system design.

Precision agriculture and the increasing automation efforts in animal husbandry requires continuous and complex monitoring of the animals. Rumen bolus sensors, which are cutting-edge pieces of technology and a rapidly developing research field, present an exceptional opportunity for monitoring the health status, physiological parameters, and estrus of the animals. The objective of this paper is to provide a comprehensive overview of the development process of a new sensor development. We address the issues of conceptual design, an overview of applicable sensor modalities, mechanical design, power supply design, applicable hardware solutions, applicable communication solutions and finally the sensor detection algorithms proved in field tests. In conclusion, we present a summary of the current opportunities in the field and provide an analysis of the foreseeable trends.

In everyday life, people often stand up and sit down. Unlike young, able-bodied individuals, older adults and those with disabilities usually stand up or sit down slowly, often pausing during the transition. It is crucial to design interfaces that accommodate these movements. Additionally, in public settings, protecting personal information is essential. Addressing these considerations, this paper presents a distance-based representation scheme for the motions of standing up and sitting down. This proposed scheme identifies both standing and sitting positions, as well as the transition process between these two states. Our scheme is based solely on the variations in distance between a sensor and the surfaces of the human body during these movements. Specifically, the proposed solution relies on distance as input, allowing for the use of a proximity sensor without the need for cameras or additional wearable sensor attachments. A single microcontroller is adequate for this purpose. Our contribution highlights that using a proximity sensor broadens the applicability of the approach while ensuring that personal information remains secure. Additionally, the scheme alleviates users' mental burden, particularly regarding privacy concerns. Extensive experiments were performed on 58 subjects, including 19 people over the age of 70, to verify the effectiveness of the proposed solution, and the results are described in detail.

This paper presents an enhanced measurement technique for evaluating embroidered transmission lines (TLs), based on a TL characterization method. The evaluation metric is the "pure" losses of the embroidered TL excluding mismatch losses. Enhanced mechanical stability and removability of embroidered samples under a test is supported by a specially designed measurement setup. Losses are used to find the effective conductivity of each embroidery pattern. Various embroidered samples are fabricated, measured, and evaluated. The repeatability of measurements and fabrication are analyzed and assessed, resulting in average deviations of 0.5 dB and 0.7 dB, respectively. A comparative evaluation of two different yarns of low and high conductivity is presented. Single and double stitching patterns for each yarn are manufactured with stitch densities of 1-7 lines/mm. For interconnection with SMA connectors, a conductive fabric contact (CFC) was selected as the finish of the TL, as a more practical interface instead of direct yarn contact (YC). The analysis of the measurements proved useful findings, such as an increase in the stitch density or the amount of yarn used does not always improve the performance; the use of double stitching greatly improves low-performance stitch densities; the effective conductivity of embroidery patterns changes with frequency; the YC interface yields more losses for medium stitch densities, but for higher stich densities, it presents an improved performance compared with the CFC interconnection.

Traditional rehabilitation training for stroke patients with ankle joint issues typically relies on the expertise of physicians. However, when confronted with complex challenges, such as online decision-making or assessing rehabilitation progress, even seasoned experts may not anticipate all potential hurdles. A novel approach is necessary-one that effectively addresses these complexities without solely leaning on expert experience. Previous studies have introduced a rehabilitation assessment method based on fuzzy neural networks. This paper proposes a novel approach, which is a VR-aided ankle rehabilitation decision-making model based on a convolutional gated recurrent neural network. This model takes various inputs, including ankle dorsiflexion range of motion, angular velocity, jerk, and motion performance scores, gathered from wearable motion inertial sensors during virtual reality rehabilitation. To overcome the challenge of limited data, data augmentation techniques are employed. This allows for the simulation of five stages of rehabilitation based on the Brunnstrom staging scale, providing tailored control parameters for virtual training scenarios suited to patients at different stages of recovery. Experiments comparing the classification performance of convolutional neural networks and long short-term memory networks were conducted. The results were compelling: the optimized convolutional gated recurrent neural network outperformed both alternatives, boasting an average accuracy of 99.16% and a Macro-F1 score of 0.9786. Importantly, it demonstrated a strong correlation (correlation coefficient r > 0.9) with the assessments made by clinical rehabilitation experts, showing its effectiveness in real-world applications.

Waste material classification is critical for efficient recycling and waste management. This study proposes a novel, low-cost material classification system based on a single, voltage-tunable Ge-on-Si photodetector operating across the visible and short-wave infrared (SWIR) spectral regions. Thanks to its tunability, the sensor is able to extract spectral information, and the system effectively distinguishes between seven different materials, including plastics, aluminum, glass, and paper. The system operates with a broadband illuminator, and material identification is obtained through the processing of the photocurrent signal at different bias voltages with classification algorithms. Here, we demonstrate the basic system functionality and near real-time classification of different waste materials.

In biological micromanipulation, cell penetration is a typical procedure that precedes cell injection or oocyte enucleation. During this procedure, cells usually undergo significant deformation, which leads to cell damage. In this paper, we focus on modeling and evaluating the cell penetration process to reduce cell deformation and stress, thereby reducing cell damage. Initially, a finite element model (FEM) is established to simulate the cell penetration process. The effectiveness of the model is then verified through visual detection and comparison of cell deformation with experimental data. Next, various mechanical responses are analyzed, considering the influence of parameters, such as the radius and shape of the injection micropipettes, material properties, and size of the cells. Finally, the relationship between the intracellular stress and the cell penetration depth of biological cells is obtained. The evaluation results will be applied to develop optimized operation plans, enhancing the efficiency and safety of the cell penetration process.

This paper proposes a novel System-in-Package (SiP) integrated architecture that incorporates Silicon-On-Insulator (SOI) switches and Surface Acoustic Wave (SAW) filters within the chip, aiming to fulfill the demands for miniaturization and multi-functionality for application in emerging wireless technologies. The proposed architecture not only reduces the integration complexity but also considers the architectural design of the integrated module, impedance matching techniques, and signal integrity for carrier aggregation (CA) technology realization. The feasibility of employing SOI switches and SAW filters based on SiP design has been validated through the trial production of a Sub-3GHz radio frequency (RF) front-end diversity receiver module. The resulting RF front-end module demonstrates exceptionally high packaging density and enhanced communication reliability, rendering it suitable for diverse applications in miniaturized RF systems.

To solve the poor performance or even failure of the cross-spectrum (CS) method in hydroacoustic weak-target passive velocimetry, a sparse Bayesian learning cross-spectrum method (SBL-CS), combining phase compensation with sparse Bayesian learning (SBL) is proposed in this paper. Firstly, the cross-correlation sound intensity is taken as the observation quantity and compensates for each frequency point of the cross-spectrum, which enables the alignment of cross-spectrum results at different frequencies. Then, the inter-correlation sound intensity of all frequencies is fused in the iterative estimation of the target velocity, verifying the proposed method's ability to suppress the background noise when performing multi-frequency processing. The simulation results show that the proposed method is still effective in estimating the target velocity when the CS method fails and that the performance of the proposed method is better than the CS method with a decrease in SNR. As verified using the SWellEx-96 sea trial dataset, the RMSE of the proposed method for surface vessel speed measurement is 0.3545 m/s, which is 46.1% less than the traditional CS method, proving the feasibility and effectiveness of the proposed SBL-CS method for the estimation of the radial speed of a passive target.

This paper presents significant improvements in the accuracy and computational efficiency of safety helmet detection within industrial environments through the optimization of the you only look once version 5 small (YOLOv5s) model structure and the enhancement of its loss function. We introduce the convolutional block attention module (CBAM) to bolster the model's sensitivity to key features, thereby enhancing detection accuracy. To address potential performance degradation issues associated with the complete intersection over union (CIoU) loss function in the original model, we implement the modified penalty-decay intersection over union (MPDIoU) loss function to achieve more stable and precise bounding box regression. Furthermore, considering the original YOLOv5s model's large parameter count, we adopt a lightweight design using the MobileNetV3 architecture and replace the original squeeze-and-excitation (SE) attention mechanism with CBAM, significantly reducing computational complexity. These improvements reduce the model's parameters from 15.7 GFLOPs to 5.7 GFLOPs while increasing the mean average precision (mAP) from 82.34% to 91.56%, demonstrating its superior performance and potential value in practical industrial applications.

With the proliferation of Internet of Things (IoT) devices and edge nodes, edge computing has taken on much of the real-time data processing and low-latency response tasks which were previously managed by cloud computing. However, edge computing often encounters challenges such as network instability and dynamic resource variations, which can lead to task interruptions or failures. To address these issues, developing a fault-tolerant scheduling mechanism is crucial to ensure that a system continues to operate efficiently even when some nodes experience failures. In this paper, we propose an innovative fault-tolerant scheduling model based on asynchronous graph reinforcement learning. This model incorporates a deep reinforcement learning framework built upon a graph neural network, allowing it to accurately capture the complex communication relationships between computing nodes. The model generates fault-tolerant scheduling actions as output, ensuring robust performance in dynamic environments. Additionally, we introduce an asynchronous model update strategy, which enhances the model's capability of real-time dynamic scheduling through multi-threaded parallel interactions with the environment and frequent model updates via running threads. The experimental results demonstrate that the proposed method outperformed the baseline algorithms in terms of quality of service (QoS) assurance and fault-tolerant scheduling capabilities.

The rapid growth of e-commerce has significantly increased demands on logistics providers, particularly in the areas of product handling and shipment. One of the most challenging and critical aspects of this process is the handling of dangerous goods (DG). This is a complex and time-intensive operation that requires safety measures and proper packaging and labelling, because mishandling DG can lead to severe injuries, property loss, and legal liability. This puts additional pressure on logistics providers to efficiently manage DG while maintaining speed and safety in the context of e-commerce. To meet this challenge, there is an urgent need to advance training programs and enhance the operational efficiency for DG handling. The use of mixed reality (MR) technology offers a promising solution. By seamlessly integrating virtual elements with real-world environments, MR has the potential to greatly improve the effectiveness and efficiency of the training of DG handling. Earlier research has examined MR in various fields, while there is still a research gap in applying MR specifically to the training of DG handling. This paper seeks to address the current research gap by presenting a novel MR model, named "MRSafe," for a training program on the safe handling of DG. The model offers users virtual experiences and comprehensive guidance to provide operational decision support.

This article examines the transformative potential of integrating reconfigurable intelligent surfaces (RISs) into sixth-generation (6G) wireless non-terrestrial networks (NTNs). The focus is on the RIS's capability to address diverse user requirements, including secure data transmission, power efficiency, extended coverage, and enhanced data rates. The paper delves into the synergy between RISs and NTNs, emphasizing key components like multiple-input multiple-output (MIMO) systems and advanced radio communications. Additionally, it highlights the crucial role of artificial intelligence (AI) and machine learning (ML) in optimizing RIS-based beamforming to solve scientific and engineering challenges while ensuring energy efficiency and sustainability in NTN operations. By positioning RISs as a key enabler in shaping the future of wireless communication systems, this research underscores their significance in unlocking the full potential of NTNs and advancing next-generation wireless communications. This paper contributes valuable insights and projections for future research directions, highlighting RISs' potential to revolutionize NTNs for 6G technologies.

As Location-Based Services (LBSs) rapidly develop, indoor localization technology is garnering significant attention as a critical component. Smartphones have become tools for indoor localization due to their highly integrated sensors, fast-evolving computational capabilities, and widespread user adoption. With the rapid advancement of smartphones, methods for smartphone-based indoor localization have increasingly attracted attention. Although there are reviews on indoor localization, there is still a lack of systematic reviews focused on smartphone-based indoor localization methods. In particular, existing reviews have not systematically analyzed smartphone-based indoor localization methods or considered the combination of smartphone sensor data with prior knowledge of the indoor environment to enhance localization performance. In this study, through systematic retrieval and analysis, the existing research was first categorized into three types to dissect the strengths and weaknesses based on the types of data sources integrated, i.e., single sensor data sources, multi-sensor data fusion, and the combination of spatial context with sensor data. Then, four key issues are discussed and the research gaps in this field are summarized. Finally, a comprehensive conclusion is provided. This paper offers a systematic reference for research and technological applications related to smartphone-based indoor localization methods.

Tropical cyclones (TC) are dynamic atmospheric phenomena featuring extreme low-pressure systems and powerful winds, known for their devastating impacts on weather and the environment. The main purpose of this paper is to consider the subtle involvement of TCs in the air quality index (AQI), focusing on aspects related to the air quality before, during and after cyclones. This research employs multimodal methods, which include meteorological data and different satellite observations. Deep learning approaches, i.e., ConvLSTM, CNN and Real-ESRGAN models, are combined with a regression model to analyze the temporal variability in the air quality associated with tropical cyclones. Deep learning models are deployed to uncover complex patterns and non-linear interdependencies between cyclones' features and the AQI to give predictive insights into the air quality fluctuations throughout the different stages of tropical cyclones. Furthermore, this study explores the aftermaths of TCs in terms of the air quality with respect to post-cyclone recovery. The findings offer an enhanced view of the role of TCs in the regional or global air quality, which will be useful for policymakers, meteorologists and environmental researchers. Utilizing a CNN for tropical cyclone (TC) classification and the extra trees regressor (ETR) for AQI prediction results in accuracy of 92.02% for the CNN and an R2 of 83.33% for the ETR. Hence, this work adds to our knowledge and enlightens us on the complex interactions between TCs and the air quality, highlighting wider public health concerns regarding climate adaptation and urban renewal.

This study aimed to show what role biomedical engineering can play in sexual health. A new concept of sexological ontology, an essential tool for building evidence-based models of sexual health, is proposed. This ontology should be based on properly validated mathematical models of sexual reactions identified using reliable measurements of physiological signals. This paper presents a review of the recommended measurement methods. Moreover, a general human sexual reaction model based on dynamic systems built at different levels of time × space × detail is presented, and the actual used modeling approaches are reviewed, referring to the introduced model. Lastly, examples of devices and computer programs designed for sexual therapy are described, indicating the need for legal regulation of their manufacturing, similar to that for other medical devices.

The vast expanses of remote onshore areas in oil-producing countries are home to a network of flow and collection pipelines that are susceptible to leaks. Most of these areas lack the infrastructure to enable the use of remote monitoring systems equipped with sensors and real-time data analysis to provide early detection of anomalies. This paper proposes a proof of concept for a monitoring system based on the Internet of Things (IoT) for real-time detection of pipeline leaks in onshore oil production fields. The proposed system, based on a thermal imaging leak detection method, informs the operator of the system's operating status via a web page. The leak detection system communicates via a Zigbee network between the IoT devices and a 4G mobile network. The results of the tests carried out show that a visual and automatic IoT-based leak detection system is possible and plausible. The proposed leak detection system enables supervisors at remote stations and field workers to monitor the operating status of pipelines via computers, tablets, or smartphones, regardless of where they are.

Torque measurement is a key task in several mechanical and structural engineering applications. Most commercial torquemeters require the shaft to be interrupted to place the sensors between the two portions of the shaft where a torque has to be measured. Contactless torquemeters based on the inverse magnetostrictive effect represent an effective alternative to conventional ones. Most known ferromagnetic materials have an inverse magnetostrictive behavior: applied stresses induce variations in their magnetic properties. This paper investigates the possibility of measuring torsional loads applied to a shaft made of ferromagnetic steel S235 through an inverse magnetostrictive torquemeter. It consists of an excitation coil that produces a time-varying electromagnetic field inside the shaft and an array of sensing coils suitably arranged around it, in which voltages are induced. First, the system is analyzed both in unloaded and loaded conditions by a Finite Element Method, investigating the influence of relative positions between the sensor and the shaft. Then, the numerical results are compared with the experimental measurements, confirming a linear characteristic of the sensor (sensitivity about 0.013 mV/Nm for the adopted experimental setup) and revealing the consistency of the model used. Since the system exploits the physical behavior of a large class of structural steel and does not require the introduction of special materials, this torquemeter may represent a reliable, economical, and easy-to-install device.

This paper proposes a Composition Piecewise Memory Polynomial (CPMP) digital predistortion model based on a Vector Switched (VS) behavioral model to address the challenges of severe nonlinearity and strong memory effects in wideband power amplifiers (PAs). To tackle this issue, two thresholds are calculated and used to segment the envelope values of the input signal according to the nonlinear distortion characteristics of the PA. In this approach, a Generalized Memory Polynomial (GMP) model is employed for the lower segment, a Memory Polynomial (MP) model is employed for the middle segment, and a higher-order GMP model is employed for the upper segment. By sharing the fundamental MP among the proposed segmented models and leveraging a design methodology that configures different cross terms, memory depths, and polynomial orders for each segment, this model achieves superior linearization performance while simultaneously reducing the computational complexity associated with model extraction. The experimental results demonstrate that the adjacent channel power ratio (ACPR) of the predistorted PA output signal using the proposed model improves from -36 dBc to -54 dBc, matching the performance of the GMP model. Furthermore, this performance is 0.5 dBc better than the Piecewise Dynamic Deviation Reduction (PDDR) and Decomposed Vector Rotation (DVR) models. Notably, the complexity of the proposed parameter extraction process is 28.8% of the DVR model, 21.79% of the GMP model, and 12.83% of the PDDR model.

The A* algorithm is an effective method for path planning; however, it has certain drawbacks, such as a high number of turns, low planning efficiency, and redundant searches. To address these issues, this paper proposes an improved bidirectional A* global path planning algorithm based on a variable neighborhood search strategy, named VNS-BA*. The new algorithm first employs an 8-11-13 neighborhood search method for node expansion. Then, the bidirectional search strategy is optimized by using the current nodes of the opposite path and the global target point, enabling the two paths to meet in the middle of the map. Finally, redundant turns are removed from the path, and cubic spline interpolation is applied to achieve local smoothing at the turns. The effectiveness of the improved algorithm was validated on different maps and compared with A* and its three derived improved versions. The simulation results indicate that VNS-BA* shows significant improvements in terms of the number of path turns, turn angles, and planning efficiency.

Objective: Pedestrian detection is very important for the environment perception and safety action of intelligent robots and autonomous driving, and is the key to ensuring the safe action of intelligent robots and auto assisted driving., Methods: In response to the characteristics of pedestrian objects occupying a small image area, diverse poses, complex scenes and severe occlusion, this paper proposes an improved pedestrian object detection method based on the YOLOv7 model, which adopts the Convolutional Block Attention Module (CBAM) attention mechanism and Deformable ConvNets v2 (DCNv2) in the two Efficient Layer Aggregation Network (ELAN) modules of the backbone feature extraction network. In addition, the detection head is replaced with a Dynamic Head (DyHead) detector head with an attention mechanism; unnecessary background information around the pedestrian object is also effectively excluded, making the model learn more concentrated feature representations., Results: Compared with the original model, the log-average miss rate of the improved YOLOv7 model is significantly reduced in both the Citypersons dataset and the INRIA dataset., Conclusions: The improved YOLOv7 model proposed in this paper achieved good performance improvement in different pedestrian detection problems. The research in this paper has important reference significance for pedestrian detection in complex scenes such as small, occluded and overlapping objects.

Luminance is the fundamental photometric quantity representing the technical meaning of brightness. It is usually measured from a distance using a matrix sensor, which is the basis of the professional instrument. However, specific technical requirements must be fulfilled to achieve accurate results. This paper considers whether modern high-resolution smartphone cameras are suitable for luminance measurements. Three cameras from high-end smartphones were evaluated on a dedicated laboratory stand. The sensors' output characteristics showed relatively good linearity of the individual R, G, and B channels. Unfortunately, the spectral sensitivities were unfavorable, as the minimum error achieved was about 17%. This device is classified outside the generally accepted quality scale for photometric instruments. The presented investigation confirmed that none of the high-resolution smartphone cameras tested was suitable for use as a universal luminance camera. However, one of the test devices can be developmental if restrictively calibrated and used only in a specialistic laboratory stand. Using a smartphone (or only its camera) for luminance measurements requires proper advanced calibration. It is possible, but it limits us to only dedicated applications. The pilot study presented in this paper will help create a suitable test stand for spectacle vision systems, e.g., virtual reality equipment.

Motion analysis is of great interest to a variety of applications, such as virtual and augmented reality and medical diagnostics. Hand movement tracking systems, in particular, are used as a human-machine interface. In most cases, these systems are based on optical or acceleration/angular speed sensors. These technologies are already well researched and used in commercial systems. In special applications, it can be advantageous to use magnetic sensors to supplement an existing system or even replace the existing sensors. The core of a motion tracking system is a localization unit. The relatively complex localization algorithms present a problem in magnetic systems, leading to a relatively large computational complexity. In this paper, a new approach for pose estimation of a kinematic chain is presented. The new algorithm is based on spatially rotating magnetic dipole sources. A spatial feature is extracted from the sensor signal, the dipole direction in which the maximum magnitude value is detected at the sensor. This is introduced as the "maximum vector". A relationship between this feature, the location vector (pointing from the magnetic source to the sensor position) and the sensor orientation is derived and subsequently exploited. By modelling the hand as a kinematic chain, the posture of the chain can be described in two ways: the knowledge about the magnetic correlations and the structure of the kinematic chain. Both are bundled in an iterative algorithm with very low complexity. The algorithm was implemented in a real-time framework and evaluated in a simulation and first laboratory tests. In tests without movement, it could be shown that there was no significant deviation between the simulated and estimated poses. In tests with periodic movements, an error in the range of 1° was found. Of particular interest here is the required computing power. This was evaluated in terms of the required computing operations and the required computing time. Initial analyses have shown that a computing time of 3 μs per joint is required on a personal computer. Lastly, the first laboratory tests basically prove the functionality of the proposed methodology.

This paper presents a rapidly exploring random tree (RRT) algorithm with an effective post waypoint shift, which is suitable for the path planning of a wheeled mobile robot under kinematic constraints. In the growth of the exploring tree, the nearest node that satisfies the kinematic constraints is selected as the parent node. Once the distance between the new node and the target is within a certain threshold, the tree growth stops and a target connection based on minimum turning radius arc is proposed to generate an initial complete random path. The most significant difference from traditional RRT-based methods is that the proposed method optimizes the path based on Dubins curves through a post waypoint shift after a random path is generated, rather than through parent node selection and rewiring during the exploring tree growth. Then, it is proved that the method can obtain an optimal path in terms of the shortest length. The optimized path has good convergence and almost does not depend on the state of the initial random path. The comparative test results show that the proposed method has significant advantages over traditional RRT-based methods in terms of the sampling point number, the tree node number, and the path node number. Subsequently, an efficient method is further proposed to avoid unknown obstacles, which utilizes the original path information and thus effectively improves the new path planning efficiency. Simulations and real-world tests are carried out to demonstrate the effectiveness of this method.

Road roughness significantly impacts vehicle safety and dynamic responses. For nonlinear suspension systems, the nonlinear characteristics often make it challenging for estimators to identify the actual road roughness accurately. This paper proposes a hybrid road roughness identification algorithm based on nonlinear auto-regressive with exogenous inputs (NARX) and a process noise adaptive square root cubature Kalman filter (PASCKF) to address this issue. Driven by vehicle acceleration data, an NARX-based road roughness identification system is constructed to mitigate the model uncertainties. Furthermore, a hybrid strategy is proposed. On the one hand, the accurate road roughness estimated by the NARX is converted into process noise covariance, enhancing the estimator's accuracy and convergence rate. Another switching strategy is proposed to optimize the non-convergence issues of the PASCKF. Finally, simulation and actual vehicle experiment data demonstrate that this approach offers superior identification accuracy and adaptability compared to the standalone SCKF algorithm.