The clustering activities of users give rise to dynamically changing ground hotspots in both space and time, posing significant challenges for providing communication services to dynamic hotspot areas. UAV swarms, with their ease of deployment and high maneuverability, offer promising solutions. However, UAVs acting as aerial base stations (BSs) are constrained by their limited computing resources and coverage range. Therefore, it is crucial to plan the flight trajectories of UAVs rationally to provide extensive coverage and maximize their resource utilization. We formulate the trajectory planning of multiple UAVs as a multi-objective optimization problem to maximize the ground user coverage, fairness, and UAV resource utilization. We propose a graph attention-enhanced deep reinforcement learning-based algorithm for the trajectory planning of multiple UAVs. The algorithm relies on a centralized training and distributed execution architecture for the flight control of multiple UAVs, enabling UAVs to fully explore the unknown area to learn the distribution of hotspot regions and perceive the optimal altitude to maximize coverage. Additionally, the algorithm incorporates a graph attention encoder to enhance UAV perception of the surrounding environment, addressing the issue of limited observation range. Extensive simulation results demonstrate that our proposed algorithm significantly outperforms baseline algorithms. The proposed algorithm achieves a user coverage rate of 95% and a coverage fairness index close to 1.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Graph Attention Enhanced Multi-UAV Trajectory Planning: A Multi-Agent Deep Reinforcement Learning Algorithm


    Additional title:

    Lect. Notes Electrical Eng.


    Contributors:
    Liu, Lianqing (editor) / Niu, Yifeng (editor) / Fu, Wenxing (editor) / Qu, Yi (editor) / Cao, Gaoxiang (author) / Yuan, Wenke (author) / He, Huasen (author) / Hou, Yunpeng (author) / Zhu, Rangang (author) / Luo, Deng (author)

    Conference:

    International Conference on Autonomous Unmanned Systems ; 2024 ; Shenyang, China September 19, 2024 - September 21, 2024



    Publication date :

    2025-03-28


    Size :

    11 pages





    Type of media :

    Article/Chapter (Book)


    Type of material :

    Electronic Resource


    Language :

    English





    Multi-Agent Trajectory Prediction with Graph Attention Isomorphism Neural Network

    Liu, Yongkang / Qi, Xuewei / Sisbot, Emrah Akin et al. | IEEE | 2022


    Parking task allocation and trajectory planning system based on multi-agent reinforcement learning

    WANG MEILING / CHEN SIYUAN / SONG WENJIE et al. | European Patent Office | 2023

    Free access


    Deep Reinforcement Learning for Image-Based Multi-Agent Coverage Path Planning

    Xu, Meng / She, Yechao / Jin, Yang et al. | IEEE | 2023