现代信息科技2025，Vol.9Issue(24)：130-137,8.DOI:10.19850/j.cnki.2096-4706.2025.24.024
基于双延迟深度确定性策略梯度的集群簇间网络拓扑控制方法
Inter-cluster Network Topology Control Method for Swarms Based on Twin Delayed Deep Deterministic Policy Gradient
摘要
Abstract
To address the cross-cluster communication interruption problem caused by relay node failures in Unmanned Aerial Vehicle(UAV)swarms,this paper proposes a topology control method based on Multi-Agent Twin Delayed Deep Deterministic Policy Gradient(MATD3).This method constructs topology and energy consumption models,and combines an improved A* algorithm with a greedy strategy to efficiently select new relay nodes.Furthermore,by designing a formation reconstruction reward function,the swarm autonomously forms a stable regular polygon formation to enhance local connectivity.Simulation results show that the proposed method significantly outperforms algorithms such as Multi-Agent Deep Deterministic Policy Gradient(MADDPG)and Multi-Agent Soft Actor-Critic(MASAC)in task efficiency,formation stability,and energy consumption,providing an effective solution for adaptive topology control of highly dynamic UAV swarm networks.关键词
无人机集群/网络连通性/中继节点/拓扑控制Key words
UAV swarm/network connectivity/relay node/topology control分类
信息技术与安全科学引用本文复制引用
杨盛,唐胜,谭潇行,戚盼,蔡圣所,沈高青,雷磊..基于双延迟深度确定性策略梯度的集群簇间网络拓扑控制方法[J].现代信息科技,2025,9(24):130-137,8.基金项目
中国自然科学基金项目(62371232) （62371232）
江苏省自然基金项目(BK20241387) （BK20241387）
中国博士后科学基金项目(2024M764237) （2024M764237）
