文献类型：期刊文献

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.[3]

机构：[1] Civil Aviation Management Institute of China, Department of General Aviation, Beijing, 100102, China; [2] School of Electronic and Information Engineering, Shenyang Aerospace University, Shenyang, 110136, China; [3] College of Robotics, Beijing Union University, Beijing, 100101, China; [4] University of Science and Technology of China, Department of Automation, Hefei, 230022, China

第一机构：Civil Aviation Management Institute of China, Department of General Aviation, Beijing, 100102, China

年份：2022

卷号：2022-July

起止页码：4996-5001

外文期刊名：Chinese Control Conference, CCC

收录：EI(收录号：20224413019186)

基金：*UDQ~ThW 1isRwork is$supported$$ by theWKHNational%HLMLQJKey(GR&DXFDWLRProgramQ &RPPofLVChinaVLRQ (Grant No. 2018AAAOlO0804), the Beijing Education Commission 6FLHQFH DQG 7HFKQRORJ\ 3URMHFW .0 .0 Science and Teclmology Project (KM201811 4 17005, KM2019114170 10), WKH =KHMLDQJ .H\ ODERUDWRU\ RI *HQHUDO $YLDWLRQ 2SHUDWLRQ WHFKQRORJ\ -'

语种：英文

外文关键词：Antennas - Computation theory - Deep learning - Game theory - Learning systems - Multi agent systems - Swarm intelligence - Unmanned aerial vehicles (UAV)

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 (UAV) 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 of UAV swarm. ? 2022 Technical Committee on Control Theory, Chinese Association of Automation.