基于合作博弈的智能集群自主聚集策略  

Strategy about autonomous aggregation of intelligent swarm based on cooperative game

作　　者：王训[1] 王兆魁[2] 张育林[1,2] WANG Xun WANG Zhaokui ZHANG Yulin(College of Aerospace Science and Engineering, National University of Defense Technology, Changsha 410073 , China School of Aerospace, Tsinghua University, Beijing 100084, China)

机构地区：[1]国防科技大学航天科学与工程学院,湖南长沙410073 [2]清华大学航天航空学院,北京100084

出　　处：《国防科技大学学报》2017年第2期146-151,共6页Journal of National University of Defense Technology

基　　金：国家863计划资助项目(2014AA7041002)

摘　　要：以无人车集群系统协同监视再入体着靶过程为任务背景,开展智能集群自组织策略相关技术研究。设计无人车集群执行再入体着靶协同监视的集群行为模式;针对协同监视过程中的集群聚集行为,提出基于合作博弈的智能集群自主聚集策略。各智能体以实现群体聚集为"合作目标",以降低自身能量消耗为"竞争目标",开展博弈;基于微粒群算法规划局部路径,最终使群体系统涌现出聚集行为。仿真实验验证了设计的自主聚集策略的有效性。Unmanned ground vehicle swarm was proposed to monitor the process of reentry vehicle landing, and then the techniques about intelligent swarm self-organizing strategy were researched. Swarm behavior patterns about unmanned vehicle swarm collaboratively monitoring the process of reentry vehicle landing were designed and the swarm self-organizing strategy based on cooperative game was presented for the swarm aggregation behavior in the process of collaborative monitoring. Each agent regards achieving swarm aggregation and saving energy as top goal and carries out the gaming. The particle swarm optimization was used to program the local path thus the aggregation behavior was realized. Simulation experiment proves the effectiveness of the autonomous aggregation strategy.

关 键 词：智能集群 无人车 再入体 合作博弈 自主聚集 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]