文献类型：会议论文

英文题名：UAV Swarm Confrontation Based on Multi-agent Deep Reinforcement Learning

作者：Wang, Zhi[1];Liu, Fan[2];Guo, Jing[2];Hong, Chen[3];Chen, Ming[3];Wang, Ershen[2];Zhao, Yunbo[4]

第一作者：Wang, Zhi

通讯作者：Hong, C[1]

机构：[1]Civil Aviat Management Inst China, Dept Gen Aviat, Beijing 100102, Peoples R China;[2]Shenyang Aerosp Univ, Sch Elect & Informat Engn, Shenyang 110136, Peoples R China;[3]Beijing Union Univ, Coll Robot, Beijing 100101, Peoples R China;[4]Univ Sci & Technol China, Dept Automat, Hefei 230022, Peoples R China

第一机构：Civil Aviat Management Inst China, Dept Gen Aviat, Beijing 100102, Peoples R China

通讯机构：[1]corresponding author), Beijing Union Univ, Coll Robot, Beijing 100101, Peoples R China.|[1141739]北京联合大学机器人学院;[11417]北京联合大学;

会议论文集：41st Chinese Control Conference (CCC)

会议日期：JUL 25-27, 2022

会议地点：Hefei, PEOPLES R CHINA

语种：英文

外文关键词：UAV Swann; Non-cooperative Game; Multi-agent; Deep Reinforcement Learning

摘要：Multi-agent deep reinforcement learning (MADRL) has attracted a tremendous amount of interest in recent years. In this paper, we introduce MADRL into the confrontation scene of Unmanned Aerial Vehicle (DAV) swarm. To analysis the dynamic game process of UAV swarm confrontation, we build two non-cooperative game models based on MADRL paradigm. By using the multi-agent deep deterministic policy gradient (MADDPG) and the centralized training with decentralized execution method, we achieve the Nash equilibrium under 5 vs. 5 UAV confrontation scenes. We also introduce multi- agent soft actor critic (MASAC) method into the UAV swarm confrontation, simulation results indicate that the MASAC-based model outperforms the :MADDPG-based model on exploring the UAV swarm combat environment, and more effectively converges to the Nash equilibrium. Our work "Will provide new insights into the confrontation ofUAV swarm.