Search Results (1,532)

Automated valet parking (AVP) is a key component of autonomous driving systems. Its functionality and reliability need to be thoroughly tested before road application. Current testing technologies are limited by insufficient scenario coverage and lack of comprehensive evaluation indices. This study proposes an AVP testing and evaluation framework using OnSite (Open Naturalistic Simulation and Testing Environment) and Unity3D platforms. Through scenario construction based on field-collected data and model reconstruction, a testing scenario library is established, complying with industry standards. A simplified kinematic model, balancing simulation accuracy and operational efficiency, is applied to describe vehicle motion. A multidimensional evaluation system is developed with completion rate as a primary index and operation performance as a secondary index, which considers both parking efficiency and accuracy. Over 500 AVP algorithms are tested on the OnSite platform, and the testing results are evaluated through the Unity3D platform. The performance of the top 10 algorithms is analyzed. The evaluation platform is compared with CARLA simulation platform and field vehicle testing. This study finds that the framework provides an effective tool for AVP testing and evaluation; a variety of high-level AVP algorithms are developed, but their flexibility in complex dynamic scenarios has limitations. Future research should focus on exploring more sophisticated learning-based algorithms to enhance AVP adaptability and performance in complex dynamic environment. Full article

Infrared detection of defects in power facilities is critical to the safe operation and fault early warning of power grids. However, conventional inspection methods have distinct limitations, such as delayed response and insufficient condition visualization. To address the pain points and technical challenges of the aforementioned inspection modes, this study proposes a deep learning network model based on multi-angle perception and Transattn feature fusion. This model can effectively improve the defect recognition ability of power facility components in complex scenarios. Firstly, a modified MAPC module is introduced, which enhances the extraction of edge contours of power facility components and detailed infrared thermal textures. Secondly, an innovative Transattn module is proposed to dynamically focus on the core component regions of power facilities. Finally, a feature fusion strategy is used to efficiently integrate the feature maps from each module, outputting component localization results and defect category information. Experimental results based on the infrared detection dataset of power facility components show that compared with classical detection models such as YOLOv10 and DDN, the proposed CMTA model achieves the best performance in all indicators: the highest mAP50 reaches 85.01%, the frame rate (FPS) is 252 frames per second, the parameter count is only 2.8 M, and it significantly shortens the fault response time of operation and maintenance personnel. Full article

In the cooperative search for dynamic targets by multiple UAVs, target uncertainty and system complexity pose significant challenges to cooperative decision-making. Multi-agent reinforcement learning (MARL) technology can be used for cooperative policy optimization, but it suffers from convergence difficulties and low policy quality in reward-sparse environments such as dynamic target search. To address this issue, this paper proposes a Multi-Potential-Field Fusion Reward Shaping MAPPO (MPRS-MAPPO) algorithm. First, three potential field functions are constructed for reward shaping: probability edge potential field, maximum probability potential field, and coverage probability sum potential field. Subsequently, an adaptive fusion weight mechanism is proposed to adjust fusion weights based on the correlation between potential field values and advantage values. Furthermore, a warm-up phase is introduced to improve training stability. Extensive experiments, including multi-scale and physical tests, demonstrate that MPRS-MAPPO significantly improves convergence speed, detection rate, and stability compared with MAPPO, MASAC, QMIX, and Scanline. Detection rates increased by 7.87–29.76%, and training uncertainty decreased by 7.43–56.36%, validating the algorithm’s robustness, scalability, and real-world applicability. Full article

In the field of orchard harvesting, this study proposes a collaborative picking strategy for dual-arm robots, aiming to improve efficiency, reduce labor burden, and achieve precise automation. The strategy combines the Multi-Agent Proximal Policy Optimization (MAPPO) algorithm with the Multi-Objective Greedy Picking Strategy (MOGPS) algorithm. By centrally training the critic network and decentralizing the actor network, the robots can autonomously learn and precisely pick in a simulated environment. To address dynamic obstacle avoidance, a dynamic collision assessment strategy is proposed, and an improved MOGPS algorithm is used to consider the distribution of fruits and the complexity of the working environment, achieving adaptive path planning. Experimental results show that the MAPPO-MOGPS algorithm optimizes the picking path by 15.11%, with a picking success rate as high as 92.3% and an average picking error of only 0.014. Additionally, physical experiments in real-world settings demonstrate the algorithm’s practical effectiveness and generalization. Full article

Cooperative pursuit using multi-UAV systems presents significant challenges in dynamic task allocation, real-time coordination, and trajectory optimization within complex environments. To address these issues, this paper proposes a reinforcement learning-based task planning framework that employs a distributed Actor–Critic architecture enhanced with bidirectional recurrent neural networks (BRNN). The pursuit–evasion scenario is modeled as a multi-agent Markov decision process, enabling each UAV to make informed decisions based on shared observations and coordinated strategies. A multi-stage reward function and a BRNN-driven communication mechanism are introduced to improve inter-agent collaboration and learning stability. Extensive simulations across various deployment scenarios, including 3-vs-1 and 5-vs-2 configurations, demonstrate that the proposed method achieves a success rate of at least 90% and reduces the average capture time by at least 19% compared to rule-based baselines, confirming its superior effectiveness, robustness, and scalability in cooperative pursuit missions. Full article

WSN coverage optimization faces two key challenges: firstly, traditional algorithms are prone to getting stuck in local optima, leading to ‘coverage holes’ in node deployment; Secondly, in dynamic scenarios (such as imbalanced energy consumption of nodes), the convergence speed of the algorithm is slow, making it difficult to maintain high coverage in real time. This study focuses on the coverage optimization problem of wireless sensor networks (WSNs) and proposes improvements to the Flamingo Search Optimization Algorithm (FSA). Specifically, the algorithm is enhanced by integrating the elite opposition-based learning strategy and the stagewise step-size control strategy, which significantly improves its overall performance. Additionally, the introduction of a cosine variation factor combined with the stagewise step-size control strategy enables the algorithm to effectively break free from local optima constraints in the later stages of iteration. The improved Flamingo Algorithm is applied to optimize the deployment strategy of sensing nodes, thereby enhancing the coverage rate of the sensor network. First, an appropriate number of sensing nodes is selected according to the target area, and the population is initialized using a chaotic sequence. Subsequently, the improved Flamingo Algorithm is adopted to optimize and solve the coverage model, with the coverage rate as the fitness function and the coordinates of all randomly distributed sensing nodes as the initial foraging positions. Next, a search for candidate foraging sources is performed to obtain the coordinates of sensing nodes with higher fitness; the coordinate components of these candidate foraging sources are further optimized through chaos theory to derive the foraging source with the highest fitness. Finally, the coordinates of the optimal foraging source are output, which correspond to the coordinate values of all sensing nodes in the target area. Experimental results show that after 100 and 200 iterations, the coverage rate of the improved Flamingo Search Optimization Algorithm is 7.48% and 5.68% higher than that of the original FSA, respectively. Furthermore, the findings indicate that, by properly configuring the Flamingo population size and the number of iterations, the improved algorithm achieves a higher coverage rate compared to other benchmark algorithms. Full article

Modern myoelectric prostheses remain difficult to control, particularly during rehabilitation, leading to high abandonment rates in favor of static devices. This highlights the need for advanced controllers that can automate some motions. This study presents an end-to-end framework coupling deep reinforcement learning with augmented reality (AR) for prosthetic actuation. A 14-degree-of-freedom hand was modeled in Blender and deployed in Unity. Two reinforcement learning agents were trained with distinct reward functions for a grasping task: (i) a discrete, Booleann reward with contact penalties and (ii) a continuous distance-based reward between joints and the target object. Each agent trained for 3 × 10⁷ timesteps at 50 Hz. The Booleann reward function performed poorly by entropy and convergence metrics, while the continuous reward function achieved success. The trained agent using the continuous reward was integrated into a dynamic AR scene, where a user controlled the prosthesis via a myoelectric armband while the grasping motion was actuated automatically. This framework demonstrates potential for assisting patients by automating certain movements to reduce initial control difficulty and improve rehabilitation outcomes. Full article

This qualitative study examines the impact of societal violence on the school climate in Arab society in Israel, focusing on teachers’ perspectives. Violence is conceptualized as an extreme, intentional form of aggression aimed at causing physical, psychological, or emotional harm. In the Israeli context, Arab society, constituting about 21% of the population, experiences disproportionately high rates of violent crime, reflecting historical marginalization, structural inequality, under-policing, and sociocultural transformations. Within schools, these societal dynamics are reported to negatively affect the learning environment, including diminished teacher motivation, concerns about teaching quality, heightened perceptions of unsafety, strained parent–school relationships, and increased parental aggression. Sixteen teachers participated in semi-structured interviews. Thematic analysis of the data revealed that financial pressures, emphasis on personal honor, and erosion of family values are perceived as key drivers of violence in the community. Teachers also reported adverse effects on students’ emotional, social, and behavioral functioning, as well as academic performance. These findings underscore the urgent need for interventions that enhance school safety, provide trauma-informed teacher training, expand psychological services, and strengthen parental collaboration. Future research should include students’ and parents’ perspectives, examine geographically diverse schools, and explore cross-cultural comparisons to better understand the educational consequences of societal violence. Full article

The Vehicular Internet of Things (V-IoT) has emerged as a cornerstone of next-generation intelligent transportation systems (ITSs), enabling applications ranging from safety-critical collision avoidance and cooperative awareness to infotainment and fleet management. These heterogeneous services impose stringent quality-of-service (QoS) demands for latency, reliability, and fairness while competing for limited and dynamically varying spectrum resources. Conventional schedulers, such as round-robin or static priority queues, lack adaptability, whereas deep reinforcement learning (DRL) solutions, though powerful, remain computationally intensive and unsuitable for real-time roadside unit (RSU) deployment. This paper proposes a lightweight and interpretable reinforcement learning (RL)-based spectrum management framework for Vehicular Internet of Things (V-IoT) networks. Two enhanced Q-Learning variants are introduced: a Value-Prioritized Action Double Q-Learning with Constraints (VPADQ-C) algorithm that enforces reliability and blocking constraints through a Constrained Markov Decision Process (CMDP) with online primal–dual optimization, and a contextual Q-Learning with Upper Confidence Bound (Q-UCB) method that integrates uncertainty-aware exploration and a Success-Rate Prior (SRP) to accelerate convergence. A Risk-Aware Heuristic baseline is also designed as a transparent, low-complexity benchmark to illustrate the interpretability–performance trade-off between rule-based and learning-driven approaches. A comprehensive simulation framework incorporating heterogeneous traffic classes, physical-layer fading, and energy-consumption dynamics is developed to evaluate throughput, delay, blocking probability, fairness, and energy efficiency. The results demonstrate that the proposed methods consistently outperform conventional Q-Learning and Double Q-Learning methods. VPADQ-C achieves the highest energy efficiency (≈$8.425\times {10}^7$ bits/J) and reduces interruption probability by over $60\%$, while Q-UCB achieves the fastest convergence (within ≈190 episodes), lowest blocking probability (≈0.0135), and lowest mean delay (≈0.351 ms). Both schemes maintain fairness near $0.364$, preserve throughput around 28 Mbps, and exhibit sublinear training-time scaling with $\mathcal{O}\left(1\right)$ per-update complexity and $\mathcal{O}\left(N^2\right)$ overall runtime growth. Scalability analysis confirms that the proposed frameworks sustain URLLC-grade latency (<0.2 ms) and reliability under dense vehicular loads, validating their suitability for real-time, large-scale V-IoT deployments. Full article

In this study, we develop and analyze an extended SEIR-type compartmental model that incorporates vaccination and treatment to describe the dynamics of acute respiratory infection transmission. The model subdivides the infectious population into several symptomatic stages and an asymptomatic class, which allows the evaluation of control strategies across different levels of infection severity. The basic reproduction number ${\mathcal{R}}_0$ is analytically derived, and its sensitivity to vaccination and treatment rates is examined to assess the impact of public health interventions on epidemic control. Numerical simulations demonstrate that the joint implementation of vaccination and treatment can markedly reduce disease prevalence and lead to infection elimination when ${\mathcal{R}}_0<1$. The results emphasize the critical role of parameter interactions in determining disease persistence and show that combining both interventions produces stronger epidemiological effects than either one alone. Machine learning techniques, specifically Support Vector Machines (SVMs), are employed to classify epidemiological outcomes and support parameter estimation. The biological markers evaluated were not effective discriminants of infection status, underscoring the importance of integrating mechanistic modeling with data-driven approaches. This combined framework enhances the understanding of epidemic dynamics and improves the predictive capacity for decision-making in public health. Full article

Efficient and stable path planning in dynamic and obstacle-dense environments, such as large-scale structure assembly measurement, is essential for improving the practicality and environmental adaptability of mobile robots in measurement and quality inspection tasks. However, traditional reinforcement learning methods often suffer from inefficient use of experience and limited capability to represent policy structures in complex dynamic scenarios. To overcome these limitations, this study proposes a method named DPDQN-TER that integrates Transformer-based sequence modeling with a multi-branch parameter policy network. The proposed method introduces a temporal-aware experience replay mechanism that employs multi-head self-attention to capture causal dependencies within state transition sequences. By dynamically weighting and sampling critical obstacle-avoidance experiences, this mechanism significantly improves learning efficiency and policy performance and stability in dynamic environments. Furthermore, a multi-branch parameter policy structure is designed to decouple continuous parameter generation tasks of different action categories into independent subnetworks, thereby reducing parameter interference and improving deployment-time efficiency. Extensive simulation experiments were conducted in both static and dynamic obstacle environments, as well as cross-environment validation. The results show that DPDQN-TER achieves higher success rates, shorter path lengths, and faster convergence compared with benchmark algorithms including Parameterized Deep Q-Network (PDQN), Multi-Pass Deep Q-Network (MPDQN), and PDQN-TER. Ablation studies further confirm that both the Transformer-enhanced replay mechanism and the multi-branch parameter policy network contribute significantly to these improvements. These findings demonstrate improved overall performance (e.g., success rate, path length, and convergence) and generalization capability of the proposed method, indicating its potential as a practical solution for autonomous navigation of mobile robots in complex industrial measurement scenarios. Full article

The Job Shop Scheduling Problem (JSP), a classical NP-hard challenge, has given rise to various complex extensions to accommodate modern manufacturing requirements. Among them, the Dynamic Flexible Job Shop Scheduling Problem (DFJSP) remains particularly challenging, due to its stochastic task arrivals, heterogeneous deadlines, and varied task types. Traditional optimization- and rule-based approaches often fail to capture these dynamics effectively. To address this gap, this study proposes a hybrid algorithm, GA-HPO PPO, tailored for the DFJSP. The method integrates genetic-algorithm–based hyperparameter optimization with proximal policy optimization to enhance learning efficiency and scheduling performance. The algorithm was trained on four datasets and evaluated on ten benchmark datasets widely adopted in DFJSP research. Comparative experiments against Double Deep Q-Network (DDQN), standard PPO, and rule-based heuristics demonstrated that GA-HPO PPO consistently achieved superior performance. Specifically, it reduced the number of overdue tasks by an average of 18.5 in 100-task scenarios and 197 in 1000-task scenarios, while maintaining a machine utilization above 67% and 28% in these respective scenarios, and limiting the makespan to within 108–114 and 506–510 time units. The model also demonstrated a 25% faster convergence rate and 30% lower variance in performance across unseen scheduling instances compared to standard PPO, confirming its robustness and generalization capability across diverse scheduling conditions. These results indicate that GA-HPO PPO provides an effective and scalable solution for the DFJSP, contributing to improved dynamic scheduling optimization in practical manufacturing environments. Full article

A surge in global connectivity has led to an increase in cyberattacks, creating a need for improved security. A promising area of research is using machine learning to detect these attacks. Traditional two-class machine learning models can be ineffective for real-time detection, as attacks often represent a minority of traffic (anomaly) and fluctuate with time. This comparative study uses an ensemble of one-class classification models. First, we employed an ensemble of autoencoders with randomly generated architectures to enhance the dynamic detection of attacks, enabling each model to learn distinct aspects of the data distribution. The term ‘dynamic’ reflects the ensemble’s superior responsiveness to different attack rates without the need for retraining, offering enhanced performance compared to a static average of individual models, which we refer to as the baseline approach. Second, for comparison with the ensemble of autoencoders, we employ an ensemble of isolation forests, which also improves dynamic attack detection. We evaluated our ensemble models using the NSL-KDD dataset, testing them without the need for retraining with varying attack ratios, and comparing the results with the baseline method. Then, we investigated the impact of training data overlap among ensemble components and its effect on the detection of extremely low attack rates. The objective is to train each model within the ensemble with the minimal amount of data necessary to detect malicious traffic across varying attack rates effectively. Based on the conclusions drawn from our initial study using the NSL-KDD dataset, we re-evaluated our strategy with a modern dataset, CIC_IoT-2023, which also achieved good performance in detecting various attack rates using an ensemble of simple autoencoder models. Finally, we have observed that when distributing normal traffic data among ensemble components with a small overlap, the results show enhanced overall performance. Full article

This paper introduces a novel framework that integrates reinforcement learning with declarative modeling and mathematical optimization for dynamic resource allocation during mass casualty incidents. Our approach leverages Mesa as an agent-based modeling library to develop a flexible and scalable simulation environment as a decision support system for emergency response. This paper addresses the challenge of efficiently allocating casualties to hospitals by combining mixed-integer linear and constraint programming while enabling a central decision-making component to adapt allocation strategies based on experience. The two-layer architecture ensures that casualty-to-hospital assignments satisfy geographical and medical constraints while optimizing resource usage. The reinforcement learning component receives feedback through agent-based simulation outcomes, using survival rates as the reward signal to guide future allocation decisions. Our experimental evaluation, using simulated emergency scenarios, shows a significant improvement in survival rates compared to traditional optimization approaches. The results indicate that the hybrid approach successfully combines the robustness of declarative modeling and the adaptability required for smart decision making in complex and dynamic emergency scenarios. Full article

The blade is a core component of the gas turbine, and blade fouling is characterized by highly concealed failure modes in the early stages and significant destructive potential in later stages. To address the lack of intelligence in early warning systems for compressor fouling, this study proposes a data-driven approach combining a digital-twin-based dynamic simulation model with the Weibull Proportional Hazards Model (WPHM) algorithm to enable reliable fault early warning. A modular design methodology was first adopted to construct a digital gas turbine model of the gas–gas combined power system on a dynamic simulation platform. High-fidelity fault simulation data were then generated to represent both healthy and faulty operating conditions. Through data governance and uncertainty quantification, key parameters influencing compressor fouling were identified. The Pearson correlation coefficient was applied to screen the most sensitive indicators, ensuring effective input selection for the prognostic model. Using historical health data from the simulation platform, the WPHM algorithm was trained to learn degradation patterns and establish a baseline failure risk model. This trained WPHM was then deployed to monitor real-time performance trends and provide early warnings for compressor blade fouling. Validation results from multi-unit simulations show that the proposed method achieves a fault warning rate of 95.0%, demonstrating its effectiveness and readiness to meet practical engineering requirements. Full article