In this work, we consider a cooperative system in which UAVs perform long-distance missions with assistance of unmanned ground vehicles (UGVs) for battery charging. We propose an autonomous landing scheme for the UAV to land on a mobile UGV with high precision by leveraging multiple-scale Quick Response (QR)-codes for different altitudes. These QR-codes support the acquisition of the relative distance and direction between the UGV and UAV. As a result, the UGV is not required to report its accurate state to the UAV. In contrast to the conventional landing algorithms based on the global positioning system (GPS), the proposed vision-based autonomous landing algorithm is available for both outdoor and GPS-denied scenarios. To cope with the challenge on the moving platform landing, a quadratic programming (QP) problem is formulated to design the velocity controller by combining a control barrier function (CBF) and a control Lyapunov function (CLF). Finally, both extensive computer simulation and a prototype of the proposed system are shown to confirm the feasibility of the proposed system and landing scheme.

There is a growing demand for indoor positioning systems (IPSs) in a wide range of applications. However, traditional solutions such as GPS face many technical challenges. In recent years, a promising alternative has been emerging, the visible light communication (VLC)-based IPS, which offers a combination of high accuracy, low cost, and energy efficiency. This article presents a comprehensive review of VLC-based IPSs, providing a tutorial-like overview of the system. It begins by comparing various positioning systems and providing background information on their inherent limitations. Experimental results have demonstrated that VLC-based systems can achieve localization accuracy to within 10 cm in controlled environments. The mechanisms of VLC-based IPSs are then discussed, including a comprehensive examination of their performance metrics and underlying assumptions. The complexity, operating range, and efficiency of VLC-based IPSs are examined by analyzing factors such as channel modeling, signal processing, and localization algorithms. To optimize VLC-based IPSs, various strategies are explored, including the design of efficient modulation schemes, the development of advanced encoding and decoding algorithms, the implementation of adaptive power control, and the application of state-of-the-art localization algorithms. In addition, system parameters are carefully examined. These include LED placement, receiver sensitivity, and transmit power. Their impact on energy efficiency and localization accuracy is highlighted. Altogether, this paper serves as a comprehensive guide to VLC IPSs, providing in-depth insights into their vast potential and the challenges that they present.

The increasing sophistication and frequency of cyber attacks necessitate automated and intelligent response mechanisms that can adapt to evolving threats. This paper presents ARCS (Adaptive Reinforcement learning for Cybersecurity Strategy), a novel framework that leverages deep reinforcement learning to optimize automated incident response strategies in cybersecurity systems. Our approach uniquely combines state representation learning of security events with a hierarchical decision-making process to map attack patterns to optimal defense measures. The framework employs a custom reward mechanism that balances incident resolution time, system stability, and defense effectiveness. Using a comprehensive dataset of 20,000 cybersecurity incidents, we demonstrate that ARCS achieves 27.3% faster incident resolution times and 31.2% higher defense effectiveness compared to traditional rule-based approaches. The framework shows particular strength in handling complex, multi-stage attacks, reducing false positive rates by 42.8% while maintaining robust system performance. Through extensive experiments, we validated that our approach can effectively generalize across different attack types and adapt to previously unseen threat patterns. The results suggest that reinforcement learning-based automation can significantly enhance cybersecurity incident response capabilities, particularly in environments requiring rapid and precise defensive actions.

Thelast decade has witnessed growing research interests in the path-planning problem for unmanned ground vehicles (UGVs) equipped with RGB-D cameras or LiDARs. However, apart from their extremely high cost, these sensors may be obscured by obstacles, which makes them impractical for many complex scenarios. Besides, most of the existing works for path planning have not considered energy efficiency for UGVs that are usually constrained by limited on-board batteries. Distinct from existing methods, this work presents a vision-based unmanned aerial vehicle (UAV)-assisted cooperative system for multiple UGVs. The proposed system harnesses the broad vision of UAV and operates in both general outdoor and global positioning system (GPS)-denied indoor environments. In sharp contrast to the conventional heuristic algorithms such as the rapidly exploring random tree (RRT) algorithm and the Dijkstra methods, the proposed energy-efficient path planning for UGVs is formulated as a non-convex optimization problem by considering the collision-aware obstacle-avoidance. More specifically, the proposed path-planning scheme contains two stages, namely the semantic segmentation that localizes UGVs and obstacles in the environment by exploiting the wide-angle camera mounted on the UAV, followed by the trajectory generation in which a disciplined convex and concave programming (DCCP) algorithm is devised for the non-convex energy-minimization problem. Extensive experiments validate the effectiveness of the UAV-UGV cooperative system and the proposed DCCP-based path-planning scheme.

Change Detection (CD) methodologies have garnered substantial attention owing to their capability to monitor alterations in geographic spaces across temporal intervals, especially with the acquisition of high-resolution Remote Sensing images. However, challenges persist due to dissimilar imaging conditions and temporal windows. Although deep-learning (DL) architectures have shown promise in addressing challenges in CD, many existing methods struggle to capture long-range dependencies and local spatial information effectively. The current CD methods rely heavily on pure CNNs and Transformers, which employ only single-pass forward propagation. This approach leads to inadequate utilization of feature information, resulting in inaccurate CD maps, particularly when discerning edges. To overcome these limitations, we propose a Transformer-based conditional generative diffusion method for CD, named TransC-GD-CD, tailored for RS data. This approach leverages the numerous sampling iterations of the DDPM, contributing to the generation of high-quality CD maps. In addition, the Frequency Cross Transformer (FCT) mechanism seamlessly amalgamates CD condition with the noise feature within the DDPM. The innovative mechanism effectively bridges diffusion noise and conditional semantic terrains. Moreover, a novel multi-type difference extraction module, named Appear-Disappear-Concat (ADC), is devised to partition the CD task to optimize both segmentation extraction and CD classification, overcoming the persistent challenge of information loss endemic to conventional CD algorithms like simple subtraction. We demonstrate the superiority of TransC-GD-CD by comparing the experiment results against various algorithms across three widely-used CD datasets, namely CDD, WHU, and LEVIR. The code for this work will be available on https://github.com/YihanWen/DDPM-based-Change-Detection .

In the field of remote sensing (RS), change detection (CD) methods are critical for analyzing the quality of images shot over various geographical areas, particularly for high-resolution images. However, there are some shortcomings of the widely used Convolutional Neural Networks (CNNs) and Transformers-based CD methods. The former is limited by its insufficient long-range modeling capabilities, while the latter is hampered by its computational complexity. Additionally, the commonly used information-fusion methods for pre- and post-change images often lead to information loss or redundancy, resulting in inaccurate edge detection. To address these issues, we propose an Iterative Mamba Diffusion Change Detection (IMDCD) approach to iteratively integrate various pieces of information and efficiently produce fine-grained CD maps. Specifically, the Swin-Mamba-Encoder (SME) within Mamba-CD (MCD) is employed as a semantic feature extractor, capable of modeling long-range relationships with linear computability. Moreover, we introduce the Variable State Space CD (VSS-CD) module, which extracts abundant CD features by training the matrix parameters within the designed State Space Change Detection (SS-CD). The computed high-dimensional CD feature is integrated into the noise predictor using a novel Global Hybrid Attention Transformer (GHAT) while low-dimensional CD features are utilized to calibrate prior CD results at each iterative step, progressively refining the generated outcomes. IMDCD exhibits a high performance across multiple datasets such as the CDD, WHU, LEVIR, and OSCD, marking a significant advancement in the methodologies within the CD field of RS. The code for this work is available on GitHub.