Search Results (2,303)

Artificial intelligence technology is profoundly reshaping the aviation field, driving the accelerated evolution of air confrontation patterns toward intelligence and autonomy. Given that experimental aircraft platforms are key means to verify intelligent air confrontation technologies, this paper—on the basis of systematically sorting out the progress of intelligent technologies in the air confrontation domain at home and abroad—first focuses on analyzing the connotation, technological evolution path, and application prospects of experimental aircraft platforms, and deeply interprets the technological breakthroughs and application practices of typical experimental platforms such as X-37B and X-62A in the field of artificial intelligence integration. Furthermore, through the analysis of three typical air confrontation projects, it reveals the four core advantages of experimental aircraft platforms in intelligent technology research: efficient iterative verification, risk reduction, promotion of capability emergence, and provision of flexible carriers. Finally, this paper focuses on constructing a technical implementation framework for the deep integration of intelligent technologies and flight tests, covering key links such as requirement analysis and environmental test design, construction of intelligent test aircraft platforms and capability generation, ground verification, and test evaluation, and summarizes various key technologies involved in the technical implementation framework. This study can provide theoretical support for the deep integration of artificial intelligence technology and the aviation field, including an engineering path from intelligent algorithm design, verification to iterative optimization, supporting the transformation of air confrontation patterns from “human-in-the-loop” to “autonomous gaming,” thereby enhancing the intelligence level and actual confrontation effectiveness in the aviation field. Full article

Perovskite materials have emerged as promising candidates for next-generation photodetectors (PDs) owing to their superior optoelectronic properties and compatibility with low-cost, low-temperature fabrication processes. Broad applicability of PDs spans diverse fields, including X-ray detection, wearable electronics, autonomous vehicles, artificial intelligence, imaging, optical communication, and biomedical sensing, offering advantages over conventional semiconductor PDs based on Si, Ge, InGaAs, and GaN. The integration of plasmonic nanostructures into perovskite-based devices has recently emerged as an effective strategy to enhance performance by amplifying light absorption near the perovskite layer. This review summarizes recent advances and design strategies for plasmonic-integrated perovskite photodetectors (Pe-PDs), with a particular emphasis on plasmonic nanopatterns and nanoparticles as viable approaches for solution-processable Pe-PDs. Full article

This study explores the use of construction robots with collaborative path planning and coordination in complex building construction tasks. Current construction processes involving robots are often fragmented due to their single-task focus, with limited research focused on employing multiple construction robots to collaboratively perform tasks. To address such a challenge, this research proposes an improved A* algorithm for global path planning and obstacle avoidance, combined with the development of a BIM-based grid map of the construction site. The leader–follower method is utilized to guide the robot group in maintaining an optimal formation, ensuring smooth collaboration during construction. The methodology includes formalizing building construction site environments into BIM-based grid maps, path planning, and obstacle avoidance, which allows robot groups to autonomously navigate and complete specific tasks such as concrete, masonry, and decoration construction. The results of this study show that the proposed approach achieves significant reductions in pathlength and operational time of approximately 9% and 10%, respectively, while maintaining safety and efficiency compared with traditional manual methods. This research demonstrates the potential of collaborative construction robot groups to enhance productivity, reduce labor costs, and provide a scalable solution for the intelligent transformation of the construction industry; extends the classical A* algorithm by incorporating obstacle density into the heuristic function; and proposes a new node simplification strategy, contributing to the literature on robot motion planning in semi-structured environments. Full article

Detecting available road space is a fundamental task for autonomous driving vehicles, requiring robust image feature extraction methods that operate reliably across diverse sensor-captured scenarios. However, existing approaches process each input independently without leveraging Accumulated Experiential Knowledge (AEK), limiting their adaptability and reliability. In order to explore the impact of AEK, we introduce MemRoadNet, a Memory-Augmented (MA) semantic segmentation framework that integrates human-inspired cognitive architectures with deep-learning models for free road space detection. Our approach combines an InternImage-XL backbone with a UPerNet decoder and a Human-like Memory Bank system implementing episodic, semantic, and working memory subsystems. The memory system stores road experiences with emotional valences based on segmentation performance, enabling intelligent retrieval and integration of relevant historical patterns during training and inference. Experimental validation on the KITTI road, Cityscapes, and R2D benchmarks demonstrates that our single-modality RGB approach achieves competitive performance with complex multimodal systems while maintaining computational efficiency and achieving top performance among single-modality methods. The MA framework represents a significant advancement in sensor-based computer vision systems, bridging computational efficiency and segmentation quality for autonomous driving applications. Full article

As artificial intelligence continues to evolve, reinforcement learning (RL) has shown remarkable potential for solving complex sequential decision problems and is now applied in diverse areas, including robotics, autonomous vehicles, and financial analytics. Among the various RL paradigms, multi-agent reinforcement learning (MARL) stands out for its ability to manage cooperative and competitive interactions within multi-entity systems. However, mainstream MARL algorithms still face critical challenges in training stability and policy generalization due to factors such as environmental non-stationarity, policy coupling, and inefficient sample utilization. To mitigate these limitations, this study introduces an enhanced algorithm named MATD3_AHD, developed by extending the MATD3 framework, which integrates TD3 and MADDPG principles. The goal is to improve the learning efficiency and overall policy effectiveness of agents operating in complex environments. The proposed method incorporates three key mechanisms: (1) an Adaptive Exploration Policy (AEP), which dynamically adjusts the perturbation magnitude based on TD error to improve both exploration capability and training stability; (2) a Hierarchical Sampling Policy (HSP), which enhances experience utilization through sample clustering and prioritized replay; and (3) a Dynamic Delayed Update (DDU), which adaptively modulates the actor update frequency based on critic network errors, thereby accelerating convergence and improving policy stability. Experiments conducted on multiple benchmark tasks within the Multi-Agent Particle Environment (MPE) demonstrate the superior performance of MATD3_AHD compared to baseline methods such as MADDPG and MATD3. The proposed MATD3_AHD algorithm outperforms baseline methods—by an average of 5% over MATD3 and 20% over MADDPG—achieving faster convergence, higher rewards, and more stable policy learning, thereby confirming its robustness and generalization capability. Full article

The production environments of coal mines are inherently complex, with interrelated disaster risks that challenge safety management. Current prediction systems struggle with fragmented data, limited mechanistic understanding, and inadequate early warnings, falling short of modern coal mine safety needs. This paper advances the thesis that artificial intelligence, including machine learning, deep learning, and Large Language Model, provides essential tools for overcoming these prediction challenges in coal mining. We review AI-based approaches for forecasting coal and gas outbursts, mine fires, water disasters, roof collapses, and dust disasters, analyzing them through technical principles, application scenarios, and empirical outcomes. The analysis clarifies how AI improves risk prediction accuracy, enhances data integration, and enables smarter decision-making for safety. By examining the five major hazards, we highlight ongoing challenges in AI implementation and outline pathways for future development, emphasizing the importance of large models and autonomous agents. Our findings support the creation of advanced AI-driven safety and early warning systems for coal mines. Full article

Object detection in high-speed and dynamic environments remains a core challenge in computer vision. Conventional frame-based cameras often suffer from motion blur and high latency, while event cameras capture brightness changes asynchronously with microsecond resolution, high dynamic range, and ultra-low latency, offering a promising alternative. Despite these advantages, existing event-based detection methods still suffer from high computational cost, limited temporal modeling, and unsatisfactory real-time performance. We present PMRVT (Parallel Attention Multilayer Perceptron Recurrent Vision Transformer), a unified framework that systematically balances early-stage efficiency, enriched spatial expressiveness, and long-horizon temporal consistency. This balance is achieved through a hybrid hierarchical backbone, a Parallel Attention Feature Fusion (PAFF) mechanism with coordinated dual-path design, and a temporal integration strategy, jointly ensuring strong accuracy and real-time performance. Extensive experiments on Gen1 and 1 Mpx datasets show that PMRVT achieves 48.7% and 48.6% mAP with inference latencies of 7.72 ms and 19.94 ms, respectively. Compared with state-of-the-art methods, PMRVT improves accuracy by 1.5 percentage points (pp) and reduces latency by 8%, striking a favorable balance between accuracy and speed and offering a reliable solution for real-time event-based vision applications. Full article

In the last decade, the incorporation of Artificial Intelligence (AI) with autonomous vehicles (AVs) has transformed transportation, mobility, and smart mobility systems. The present study provides a systematic review of global trends, applications, and challenges at the intersection of AI, including Machine Learning (ML), Deep Learning (DL), and autonomous vehicle technologies. Using data extracted from Clarivate Analytics’ Web of Science Core Collection and a set of specific keywords related to both AI and autonomous (electric) vehicles, this paper identifies the themes presented in the scientific literature using thematic maps and thematic map evolution analysis. Furthermore, the research topics are identified using both thematic maps, as well as Latent Dirichlet Allocation (LDA) and BERTopic, offering a more faceted insight into the research field as LDA enables the probabilistic discovery of high-level research themes, while BERTopic, based on transformer-based language models, captures deeper semantic patterns and emerging topics over time. This approach offers richer insights into the systematic review analysis, while comparison in the results obtained through the various methods considered leads to a better overview of the themes associated with the field of AI in autonomous vehicles. As a result, a strong correspondence can be observed between core topics, such as object detection, driving models, control, safety, cybersecurity and system vulnerabilities. The findings offer a roadmap for researchers and industry practitioners, by outlining critical gaps and discussing the opportunities for future exploration. Full article

Smart manufacturing demands adaptive, scalable, and human-centric solutions for predictive maintenance. This paper introduces the concept of Agentic AI, a paradigm that extends beyond traditional multi-agent systems and collaborative AI by emphasizing agency: the ability of AI entities to act autonomously, coordinate proactively, and remain accountable under human oversight. Through federated learning, edge computing, and distributed intelligence, the proposed framework enables intentional, goal-oriented monitoring agents to form self-organizing predictive maintenance ecosystems. Validated in a ceramic manufacturing facility, the system achieved 94% predictive accuracy, a 67% reduction in false positives, and a 43% decrease in unplanned downtime. Economic analysis confirmed financial viability with a 1.6-year payback period and a €447,300 NPV over five years. The framework also embeds explainable AI and trust calibration mechanisms, ensuring transparency and safe human–machine collaboration. These results demonstrate that Agentic AI provides both conceptual and practical pathways for transitioning from reactive monitoring to resilient, autonomous, and human-centered industrial intelligence. Full article

Unmanned aerial vehicle (UAV) remote sensing is critical in assessing post-earthquake building damage. However, intelligent disaster assessment via remote sensing faces formidable challenges from complex backgrounds, substantial scale variations in targets, and diverse spatial disaster dynamics. To address these issues, we propose UEBNet, a high-precision post-earthquake building instance segmentation model that systematically enhances damage recognition by integrating three key modules. Firstly, the Depthwise Separable Convolutional Block Attention Module suppresses background noise that visually resembles damaged structures. This is achieved by expanding the receptive field using multi-scale pooling and dilated convolutions. Secondly, the Multi-feature Fusion Module generates scale-robust feature representations for damaged buildings with significant size differences by processing feature streams from different receptive fields in parallel. Finally, the Adaptive Multi-Scale Interaction Module accurately reconstructs the irregular contours of damaged buildings through an advanced feature alignment mechanism. Extensive experiments were conducted using UAV imagery collected after the Ms 6.8 earthquake in Tingri County, Tibet Autonomous Region, China, on 7 January 2025, and the Ms 6.2 earthquake in Jishishan County, Gansu Province, China, on 18 December 2023. Results indicate that UEBNet enhances segmentation mean Average Precision (${mAP}^{seg}$) and bounding box mean Average Precision (${mAP}^{box}$) by 3.09% and 2.20%, respectively, with equivalent improvements of 2.65% in F1-score and 1.54% in overall accuracy, outperforming state-of-the-art instance segmentation models. These results demonstrate the effectiveness and reliability of UEBNet in accurately segmenting earthquake-damaged buildings in complex post-disaster scenarios, offering valuable support for emergency response and disaster relief. Full article

Modern active distribution networks are increasingly characterized by high complexity, uncertainty, and distributed clustering, posing challenges for traditional model-based simulations in capturing nonlinear dynamics and stochastic variations. This study develops a data–model hybrid-driven simulation architecture that integrates a Federated Evolutionary Monte Carlo Optimization (FEMCO) algorithm for distribution network optimization. The model-driven module employs spectral clustering to decompose the network into multiple autonomous subsystems and performs distributed reconstruction through gradient descent. The data-driven module, built upon Long Short-Term Memory (LSTM) networks, learns temporal dependencies between load curves and operational parameters to enhance predictive accuracy. These two modules are fused via a Random Forest ensemble, while FEMCO jointly leverages Monte Carlo global sampling, Federated Learning-based distributed training, and Genetic Algorithm-driven evolutionary optimization. Simulation studies on the IEEE 33 bus distribution system demonstrate that the proposed framework reduces power losses by 25–45% and voltage deviations by 75–85% compared with conventional Genetic Algorithm and Monte Carlo approaches. The results confirm that the proposed hybrid architecture effectively improves convergence stability, optimization precision, and adaptability, providing a scalable solution for the intelligent operation and distributed control of modern power distribution systems. Full article

This exploratory proof-of-concept study investigated the simultaneous deployment of autonomous, persona-driven Artificial Intelligence (AI) agents across multiple social media platforms using the ElizaOS framework. We developed three platform-specific agents with seven-layer character architectures and deployed them on Twitter/X, Discord, and Telegram for 18 days. The system processed 5389 interactions while gathering feedback from 28 volunteer participants. Addressing three research questions, we found that: (1) automation effectiveness was platform-dependent, with direct support platforms (Telegram, Discord) rated more useful than broadcast-oriented Twitter/X; (2) character design impact depended primarily on platform-persona alignment rather than architectural sophistication; and (3) technical performance showed platform-specific patterns, with median storage times ranging from 9.0 milliseconds (Twitter/X) to 61.5 milliseconds (Telegram) and high variability across all platforms. A notable finding was what we term the “Discord Paradox”—high quality ratings (4.05/5) but lowest preference (8.7%), suggesting platform familiarity and accessibility influence adoption more than agent quality. While the deployment demonstrated technical feasibility and revealed distinct user dynamics across platforms, the findings indicate that platform-specific optimization may be more effective than universal approaches. This exploratory study advances understanding of multi-platform agent deployment for marketing automation, identifying behavioral patterns and platform-specific dynamics that offer testable hypotheses for future systematic research. Full article

The Artificial Intelligence of Things (AIoT) is rapidly evolving from basic connectivity to intelligent perception, reasoning, and decision making across domains such as healthcare, manufacturing, transportation, and smart cities. Multimodal generative AI (GAI) and digital twins (DTs) provide complementary solutions. DTs deliver high-fidelity virtual replicas for real-time monitoring, simulation, and optimization with GAI enhancing cognition, cross-modal understanding, and the generation of synthetic data. This survey presents a comprehensive overview of DT–GAI integration in the AIoT. We review the foundations of DTs and multimodal GAI and highlight their complementary roles. We further introduce the Sense–Map–Generate–Act (SMGA) framework, illustrating their interaction through the SMGA loop. We discuss key enabling technologies, including multimodal data fusion, dynamic DT evolution, and cloud–edge–end collaboration. Representative application scenarios, including smart manufacturing, smart cities, autonomous driving, and healthcare, are examined to demonstrate their practical impact. Finally, we outline open challenges, including efficiency, reliability, privacy, and standardization, and we provide directions for future research toward sustainable, trustworthy, and intelligent AIoT systems. Full article

Federated learning (FL) enables distributed model training across sensor-equipped edge devices while preserving data privacy. However, its performance is often hindered by statistical heterogeneity among clients and system heterogeneity in dynamic wireless networks. To address these challenges, we propose TCS-FEEL, a topology-aware client selection framework that jointly considers user distribution, device-to-device (D2D) communication, and statistical similarity of client data. The proposed approach integrates randomized client sampling with an adaptive tree-based communication structure, where user devices not only participate in local model training but also serve as relays to exploit efficient D2D transmission. TCS-FEEL is particularly suited for sensor-driven edge intelligence scenarios such as autonomous driving, smart city monitoring, and the Industrial IoT, where real-time performance and efficient resource utilization are crucial. Extensive experiments on MNIST and CIFAR-10 under various non-IID data distributions and mobility settings demonstrated that TCS-FEEL consistently reduced the number of training rounds and shortened per-round wall-clock time compared with existing baselines while maintaining model accuracy. These results highlight that integrating topology control with client selection provides an effective solution for accelerating privacy-preserving and resource-efficient FL in dynamic, sensor-rich edge environments. Full article