Search Results (69)

Coordinating Multi-Unmanned Surface Vehicle (USV) swarms in complex, adversarial maritime environments is a significant challenge, as existing multi-agent reinforcement learning (MARL) methods often fail to capture intricate spatiotemporal dependencies, leading to suboptimal policies. To address this, we propose Adv-TransAC, a novel Spatio-Temporal Meta-Reinforcement Learning framework. Its core innovation is a hybrid GAT-transformer architecture that decouples spatial and temporal reasoning: a Graph Attention Network (GAT) models instantaneous tactical formations, while a transformer analyzes their temporal evolution to infer intent. This is combined with an adversarial meta-learning mechanism to enable rapid adaptation to opponent tactics. In high-fidelity escort and defense simulations, Adv-TransAC significantly outperforms state-of-the-art MARL baselines in task success rate and policy robustness. The learned policies demonstrate the emergence of complex cooperative behaviors, such as intelligent risk-aware coordination and proactive interception maneuvers. The framework’s practicality is further validated by a communication-efficient federated optimization architecture. By effectively modeling spatiotemporal dynamics and enabling rapid adaptation, Adv-TransAC provides a powerful solution that moves beyond reactive decision-making, establishing a strong foundation for next-generation, intelligent maritime platforms. Full article

Background: The increasing deployment of unmanned aerial vehicles (UAVs) for logistics in smart cities presents pressing challenges related to identity spoofing, unauthorized payload transport, and airspace security. Existing drone defense systems (DDSs) struggle to verify both drone identity and payload authenticity in real time, while blockchain-assisted solutions are often hindered by high latency and limited scalability. Methods: To address these challenges, we propose iBANDA, a blockchain- and AI-assisted DDS framework. The system integrates a lightweight You Only Look Once 5 small (YOLOv5s) object detection model with a Snowball-based Proof-of-Stake consensus mechanism to enable dual-layer authentication of drones and their attached payloads. Authentication processes are coordinated through an edge-deployable decentralized application (DApp). Results: The experimental evaluation demonstrates that iBANDA achieves a mean average precision of 99.5%, recall of 100%, and an F1-score of 99.8% at an inference time of 0.021 s, validating its suitability for edge devices. Blockchain integration achieved an average network latency of 97.7 ms and an end-to-end transaction latency of 1.6 s, outperforming Goerli, Sepolia, and Polygon Mumbai testnets in scalability and throughput. Adversarial testing further confirmed resilience to Sybil attacks and GPS spoofing, maintaining a false acceptance rate below 2.5% and continuity above 96%. Conclusions: iBANDA demonstrates that combining AI-based visual detection with blockchain consensus provides a secure, low-latency, and scalable authentication mechanism for UAV-based logistics. Future work will explore large-scale deployment in heterogeneous UAV networks and formal verification of smart contracts to strengthen resilience in safety-critical environments. Full article

The accelerated evolution of remote sensing technology has intensified the demand for real-time tile map generation, highlighting the limitations of conventional mapping approaches that rely on manual cartography and field surveys. To address the critical need for rapid cartographic updates, this study presents a novel multi-stage generative framework that synergistically integrates Generative Adversarial Networks (GANs) with Diffusion Denoising Models (DMs) for high-fidelity map generation from remote sensing imagery. Specifically, our proposed architecture first employs GANs for rapid preliminary map generation, followed by a cascaded diffusion process that progressively refines topological details and spatial accuracy through iterative denoising. Furthermore, we propose a hybrid attention mechanism that strategically combines channel-wise feature recalibration with coordinate-aware spatial modulation, enabling the enhanced discrimination of geographic features under challenging conditions involving edge ambiguity and environmental noise. Quantitative evaluations demonstrate that our method significantly surpasses established baselines in both structural consistency and geometric fidelity. This framework establishes an operational paradigm for automated, rapid-response cartography, demonstrating a particular utility in time-sensitive applications including disaster impact assessment, unmapped terrain documentation, and dynamic environmental surveillance. Full article

Fall detection is a critical task in healthcare monitoring systems, especially for elderly populations, for whom timely intervention can significantly reduce morbidity and mortality. This study proposes a privacy-preserving and scalable fall-detection framework that integrates federated learning (FL) with transfer learning (TL) to train deep learning models across decentralized data sources without compromising user privacy. The pipeline begins with data acquisition, in which annotated video-based fall-detection datasets formatted in YOLO are used to extract image crops of human subjects. These images are then preprocessed, resized, normalized, and relabeled into binary classes (fall vs. non-fall). A stratified 80/10/10 split ensures balanced training, validation, and testing. To simulate real-world federated environments, the training data is partitioned across multiple clients, each performing local training using pretrained CNN models including MobileNetV2, VGG16, EfficientNetB0, and ResNet50. Two FL topologies are implemented: a centralized server-coordinated scheme and a ring-based decentralized topology. During each round, only model weights are shared, and federated averaging (FedAvg) is applied for global aggregation. The models were trained using three random seeds to ensure result robustness and stability across varying data partitions. Among all configurations, decentralized MobileNetV2 achieved the best results, with a mean test accuracy of 0.9927, F1-score of 0.9917, and average training time of 111.17 s per round. These findings highlight the model’s strong generalization, low computational burden, and suitability for edge deployment. Future work will extend evaluation to external datasets and address issues such as client drift and adversarial robustness in federated environments. Full article

Modern electronic information system software is becoming increasingly complex, making manual debugging prohibitively expensive and necessitating automated fault localization (FL) methods to prioritize suspicious code segments. While Single-Fault Localization (SFL) methods, such as spectrum-based fault localization (SBFL) and Deep Learning-Based Fault Localization (DLFL), have demonstrated promising results in localizing individual faults, extending these methods to multiple-fault scenarios remains challenging. Deep Learning–Based Fault Localization (DLFL) methods combine metamorphic testing and clustering to locate multiple faults without relying on test oracles. However, these approaches suffer from a severe class imbalance problem: the number of failed cases (the minority class) is far smaller than that of passed cases (the majority class). To address this issue, we propose MetaGAN: Metamorphic GAN-based Augmentation for Improving Deep Learning-based Multiple-Fault Localization Without Test Oracles. MetaGAN is a novel method that integrates Metamorphic Testing (MT), clustering-based fault isolation, and Generative Adversarial Networks (GANs). The method first utilizes MT to gather information from failed Metamorphic Test Groups (MTGs) and extracts metamorphic features that capture the underlying failure causes to represent each failed MTG; then, these features are used to cluster the failed MTGs into several groups, with each group forming an independent single-fault debugging session; finally, in each session, data augmentation is performed by combining MT with a GAN model to generate failed test cases (the minority class) until their number matches that of passed test cases (the majority class), thereby balancing the dataset for precise DLFL-based fault localization and enabling parallel debugging of multiple faults. Extensive experimental validation on an expanded open-source benchmark shows that, compared with the baseline MetaMDFL, MetaGAN significantly improves fault localization accuracy, particularly in parallel multiple-fault scenarios. Specifically, MetaGAN achieves significant improvements in both the EXAM and the rank metrics, with EXAM showing the highest improvement of $7.81\%$, the rank showing the highest improvement of $12.71\%$, and the top-$N\%$ showing the highest improvement of $9.62\%$. This method, through coordinated dynamic feature extraction, adaptive data augmentation, and distributed collaborative debugging, provides a scalable solution for complex systems where test oracles are unavailable, thereby advancing state-of-the-art methods. Full article

The increasing integration of photovoltaic (PV) systems into smart grids introduces new cybersecurity vulnerabilities, particularly against cyber-physical attacks that can manipulate grid operations and disrupt renewable energy generation. This paper proposes a multi-layered cyber-resilient PV optimization framework, leveraging digital twin-based deception, reinforcement learning-driven cyber defense, and blockchain authentication to enhance grid security and operational efficiency. A deceptive cyber-defense mechanism is developed using digital twin technology to mislead adversaries, dynamically generating synthetic PV operational data to divert attack focus away from real assets. A deep reinforcement learning (DRL)-based defense model optimizes adaptive attack mitigation strategies, ensuring real-time response to evolving cyber threats. Blockchain authentication is incorporated to prevent unauthorized data manipulation and secure system integrity. The proposed framework is modeled as a multi-objective optimization problem, balancing attack diversion efficiency, system resilience, computational overhead, and energy dispatch efficiency. A non-dominated sorting genetic algorithm (NSGA-III) is employed to achieve Pareto-optimal solutions, ensuring high system resilience while minimizing computational burdens. Extensive case studies on a realistic PV-integrated smart grid test system demonstrate that the framework achieves an attack diversion efficiency of up to 94.2%, improves cyberattack detection rates to 98.5%, and maintains an energy dispatch efficiency above 96.2%, even under coordinated cyber threats. Furthermore, computational overhead is analyzed to ensure that security interventions do not impose excessive delays on grid operation. The results validate that digital twin-based deception, reinforcement learning, and blockchain authentication can significantly enhance cyber-resilience in PV-integrated smart grids. This research provides a scalable and adaptive cybersecurity framework that can be applied to future renewable energy systems, ensuring grid security, operational stability, and sustainable energy management under adversarial conditions. Full article

Practical large-scale multiple unmanned aerial vehicle (multi-UAV) networks are susceptible to multiple potential points of vulnerability, such as hardware failures or adversarial attacks. Existing resilient multi-dimensional coordination control algorithms in multi-UAV networks are rather costly in the computation of a safe point and rely on an assumption of the maximum number of adversarial nodes in the multi-UAV network or neighborhood. In this paper, a dynamic trusted convex hull method is proposed to filter received states in multi-dimensional space without requiring assumptions about the maximum adversaries. Based on the proposed method, a distributed local control protocol is designed with lower computational complexity and higher tolerance of adversarial nodes. Sufficient and necessary graph-theoretic conditions are obtained to achieve resilient multi-dimensional consensus and containment control despite adversarial nodes’ behaviors. The theoretical results are validated through simulations. Full article

As B5G systems are evolving toward 6G, their coordination increasingly relies on AI-driven automation and orchestration actions, a process that is characterized as cognition. Therefore, a 6G system, through this cognitive process, acts as an intent-handling entity that comprehends sophisticated intent semantics from the users/tenants and calculates the ideal goal state for the specific intent, organizing the necessary adaptation actions that are needed for the transition of the system into that state. However, the use of cognitive-driven AI models to coordinate the purposes of a 6G system creates new risks, as a new surface of attack is born, where the whole 6G system operation may be maliciously affected by adversarial attacks within the user-intents. Focusing on this challenge, this paper realizes a prototype cognitive coordinator for 6G trustworthiness provision and investigates its adversarial robustness for different BERT-based quantification models, which are used for realizing the 6G cognitive system. Full article

Federated Learning (FL) systems are increasingly vulnerable to data poisoning attacks, in which malicious clients attempt to manipulate their training data in order to compromise the corresponding machine learning model. Existing detection techniques rely mostly on identifying clients who provide weight updates that significantly diverge from the average across multiple training rounds. In this work, we propose a Clique-Based Detection Framework (CBDF) that focuses on similarity patterns between client updates instead of their deviation. Specifically, we make use of the Euclidean distance to measure similarity between the weight update vectors of different clients over training iterations. Clients that provide consistently similar weight updates and exceed a predefined threshold are flagged as potential adversaries. Therefore, this method detects the coordination patterns of the attackers and uses them to strengthen FL systems against sophisticated, coordinated data poisoning attacks. We validate the effectiveness of this approach through extensive experimental evaluation. Moreover, we provide suggestions regarding fine-tuning hyperparameters to maximize the performance of the detection method. This approach represents a novel advancement in protecting FL models from malicious interference. Full article

With the proposal of carbon peaking and carbon neutrality goals, the proportion of distributed renewable energy generation in active distribution networks (ADNs) has been continuously increasing. While this has effectively reduced greenhouse gas emissions, it has also given rise to power quality issues such as excessive or insufficient voltage amplitudes. To effectively address this problem, this paper proposes a multi-resource coordinated dynamic reactive power–voltage coordination optimization method. Firstly, an improved Generative Convolutional Adversarial Network (GCAN) is used to generate typical wind and solar power output scenarios. Based on these generated typical scenarios, a voltage control model for ADNs is established with the objective of minimizing voltage fluctuations, fully exploiting the dynamic reactive power regulation resources within the ADN. In view of the non-convex and nonlinear characteristics of the model, an improved Gray Wolf Optimizer (GWO) algorithm is employed for model optimization and solution seeking. Finally, the effectiveness and feasibility of the proposed method are demonstrated through simulations using modified IEEE-33-bus and IEEE-69-bus test systems. Full article

Autonomous vessels are becoming paramount to ocean transportation, while they also face complex risks in dynamic marine environments. Machine learning plays a crucial role in enhancing maritime safety by leveraging its data analysis and predictive capabilities. However, there has been no review grounded in bibliometric analysis in this field. To explore the research evolution and knowledge frontier in the field of maritime safety for autonomous shipping, a bibliometric analysis was conducted using 719 publications from the Web of Science database, covering the period from 2000 up to May 2024. This study utilized VOSviewer, alongside traditional literature analysis methods, to construct a knowledge network map and perform cluster analysis, thereby identifying research hotspots, evolution trends, and emerging knowledge frontiers. The findings reveal a robust cooperative network among journals, researchers, research institutions, and countries or regions, underscoring the interdisciplinary nature of this research domain. Through the review, we found that maritime safety machine learning methods are evolving toward a systematic and comprehensive direction, and the integration with AI and human interaction may be the next bellwether. Future research will concentrate on three main areas: evolving safety objectives towards proactive management and autonomous coordination, developing advanced safety technologies, such as bio-inspired sensors, quantum machine learning, and self-healing systems, and enhancing decision-making with machine learning algorithms such as generative adversarial networks (GANs), hierarchical reinforcement learning (HRL), and federated learning. By visualizing collaborative networks, analyzing evolutionary trends, and identifying research hotspots, this study lays a groundwork for pioneering advancements and sets a visionary angle for the future of safety in autonomous shipping. Moreover, it also facilitates partnerships between industry and academia, making for concerted efforts in the domain of USVs. Full article

To address the challenges of dynamic adversarial scenario modeling distortion, insufficient cross-institutional data privacy protection, and simplistic evaluation systems in collegiate basketball tactical education, this study proposes and validates an immersive instructional system integrating digital twin and federated learning technologies. The four-tier architecture (sensing layer, digital twin layer, federated layer, and interaction layer) synthesizes multimodal data (motion trajectories and physiological signals) with Multi-Agent Reinforcement Learning (MARL) to enable virtual–physical integrated tactical simulation and real-time error correction. Experimental results demonstrate that the experimental group achieved 35.2% higher tactical execution accuracy (TEA) (p < 0.01), 1.8 s faster decision making (p < 0.05), and 47% improved team coordination efficiency compared to the controls. The hierarchical federated learning framework (trajectory ε = 0.8; physiology ε = 0.3) maintained model precision loss at 2.4% while optimizing communication efficiency by 23%, ensuring privacy preservation. A novel three-dimensional “Skill–Creativity–Load” evaluation system revealed a 22% increase in unconventional tactical applications (p = 0.013) through the Tactical Creativity Index (TCI). By implementing lightweight federated architecture with dynamic cognitive offloading mechanisms, the system enables resource-constrained institutions to achieve 87% of the pedagogical effectiveness observed in elite programs, offering an innovative solution to reconcile educational equity with technological ethics. Future research should focus on long-term skill transfer, multimodal adaptive learning, and ethical framework development to advance intelligent sports education from efficiency-oriented paradigms to competency-based transformation. Full article

In modern naval confrontation systems, adversarial underwater unmanned vehicles (UUVs) pose significant challenges, which are deployed on unmanned aerial vehicles (UAVs) due to their inherent mobility and positional uncertainty. Effective neutralization threats demand sophisticated coordination strategies between distributed agents under partial observability. This paper proposes a novel Knowledge-Enhanced Multi-Agent Deep Reinforcement Learning (MADRL) framework for coordinating UAV swarms against adversarial UUVs in asymmetric confrontation scenarios, specifically addressing three operational modes: area surveillance, summoned interception, and coordinated countermeasures. Our framework introduces three key innovations: (1) a probabilistic adversarial model integrating prior intelligence and real-time UAV sensor data to predict underwater trajectories; (2) a Multi-Agent Double Soft Actor–Critic (MADSAC) algorithm, addressing Red team coordination challenges. Experimental validation demonstrates superior performance over baseline methods in Blue target detection efficiency (38.7% improvement) and successful neutralization rate (52.1% increase), validated across escalating confrontation scenarios. Full article

Terrestrial and satellite communications, tactical data links, positioning, navigation, and timing (PNT), as well as distributed sensing will continue to require precise timing and the ability to synchronize and disseminate time effectively. However, the supply of space-qualified clocks that meet Global Navigation Satellite Systems (GNSS)-level performance standards is limited. As the awareness of potential disruptions to GNSS due to adversarial actions grows, the current reliance on GNSS-level timing appears costly and outdated. This is especially relevant given the benefits of developing robust and stable time scale references in orbit, especially as various alternatives to GNSS are being explored. The onboard realization of clock ensembles is particularly promising for applications such as those providing the on-demand dissemination of a reference time scale for navigation services via a proliferated Low-Earth Orbit (pLEO) constellation. This article investigates potential inter-satellite network architectures for coordinating time and frequency across pLEO platforms. These architectures dynamically allocate radio resources for clock data transport based on the requirements for pLEO time scale formations. Additionally, this work proposes a model-based control system for wireless networked timekeeping systems. It envisions the optimal placement of critical information concerning the implicit ensemble mean (IEM) estimation across a multi-platform clock ensemble, which can offer better stability than relying on any single ensemble member. This approach aims to reduce data traffic flexibly. By making the IEM estimation sensor more intelligent and running it on the anchor platform while also optimizing the steering of remote frequency standards on participating platforms, the networked control system can better predict the future behavior of local reference clocks paired with low-noise oscillators. This system would then send precise IEM estimation information at critical moments to ensure a common pLEO time scale is realized across all participating platforms. Clock steering is essential for establishing these time scales, and the effectiveness of the realization depends on the selected control intervals and steering techniques. To enhance performance reliability beyond what the existing Linear Quadratic Gaussian (LQG) control technique can provide, the minimal-cost-variance (MCV) control theory is proposed for clock steering operations. The steering process enabled by the MCV control technique significantly impacts the overall performance reliability of the time scale, which is generated by the onboard ensemble of compact, lightweight, and low-power clocks. This is achieved by minimizing the variance of the chi-squared random performance of LQG control while maintaining a constraint on its mean. Full article

To address the challenges posed by complex backgrounds and the low occurrence in photovoltaic cell images captured by industrial sensors, we propose a novel defect detection method: MRA-YOLOv8. First, a multi-branch coordinate attention network (MBCANet) is introduced into the backbone. The coordinate attention network (CANet) is incorporated to mitigate the noise impact of background information on the detection task, and multiple branches are employed to enhance the model’s feature extraction capability. Second, we integrate a multi-path feature extraction module, ResBlock, into the neck. This module provides finer-grained multi-scale features, improving feature extraction from complex backgrounds and enhancing the model’s robustness. Finally, we implement alpha-minimum point distance-based IoU (AMPDIoU) to the head. This loss function enhances the accuracy and robustness of small object detection by integrating minimum point distance-based IoU (MPDIoU) and Alpha-IoU methods. The results demonstrate that MRA-YOLOv8 outperforms other mainstream methods in detection performance. On the photovoltaic electroluminescence anomaly detection (PVEL-AD) dataset, the proposed method achieves a mAP₅₀ of 91.7%, representing an improvement of 3.1% over YOLOv8 and 16.1% over detection transformer (DETR). On the SPDI dataset, our method achieves a mAP₅₀ of 69.3%, showing a 2.1% improvement over YOLOv8 and a 6.6% improvement over DETR. The proposed MRA-YOLOv8 also exhibits great deployment potential. It can be effectively integrated with drone-based inspection systems, allowing for efficient and accurate PV plant inspections. Moreover, to tackle the issue of data imbalance, we propose generating synthetic defect data via generative adversarial networks (GANs), which can supplement the limited defect samples and improve the model’s generalization ability. Full article