文献类型：期刊文献

中文题名：Counterfactual baseline-based MAPPO for asymmetric UAV swarm confrontation game

作者：Ershen WANG[1,2];Zeqi TONG[1];Chen HONG[3,4];Xiaotong WU[1];Mingming XIAO[5];Chang LIU[4];Jihao CHEN[1]

第一作者：Ershen WANG

机构：[1]School of Electronic and Information Engineering,Shenyang Aerospace University,Shenyang 110136,China;[2]Key Laboratory of Technology and Equipment of Tianjin Urban Air Transportation System,Civil Aviation University of China,Tianjin 300300,China;[3]Multi-Agent Systems Research Centre,Beijing Union University,Beijing 100101,China;[4]College of Robotics,Beijing Union University,Beijing 100101,China;[5]College of Smart City,Beijing Union University,Beijing 100101,China

第一机构：School of Electronic and Information Engineering,Shenyang Aerospace University,Shenyang 110136,China

年份：2026

卷号：69

期号：2

起止页码：513-514

中文期刊名：Science China(Information Sciences)

外文期刊名：中国科学(信息科学)(英文版)

基金：supported by National Key R&D Program of China(Grant No.2018AAA0100804);Basic Science Center Program of National Natural Science Foundation of China(Grant No.62388101);National Natural Science Foundation of China(Grant No.62173237);Aeronautical Science Foundation of China(Grant No.20240055054001);Open Fund of Key Laboratory of Technology and Equipment of Tianjin Urban Air Transportation System(Grant No.TJKL-UAM-202305);Joint Fund of Ministry of Natural Resources Key Laboratory of Spatiotemporal Perception and Intelligent Processing(Grant No.232203);Applied Basic Research Programs of Liaoning Province(Grant No.2025JH2/101300011)。

语种：英文

中文关键词：Unmanned Aerial Vehicle;Multi Agent Deep Reinforcement Learning;CB MAPPO;internal cooperation;make optimal decision;Swarm Confrontation;unmanned aerial vehicle uav manufacturing;Counterfactual Baseline;

摘要：With the rapid development of unmanned aerial vehicle(UAV)manufacturing technology and the increasing complexity of the UAV working environment,UAV swarms are bound to thrive in future military and civilian tasks.Effective internal cooperation across UAVs is crucial for accomplishing the mission successfully in the UAV swarm confrontation game[1].How to make optimal decision in asymmetric UAV swarm confrontation tasks is still an open issue[2,3].In this study,we propose a novel multi-agent deep reinforcement learning method named counterfactual baselinebased MAPPO(CB-MAPPO)to tackle the decision-making issues of asymmetric UAV swarm confrontation missions.Our contributions are summarized as follows.