首页 | 本学科首页   官方微博 | 高级检索  
     检索      

一种未知动态环境下异构无人机集群分布式联盟形成方法
引用本文:郑红星,郭继峰,谢旭东,颜鹏.一种未知动态环境下异构无人机集群分布式联盟形成方法[J].宇航学报,2022,43(2):189-197.
作者姓名:郑红星  郭继峰  谢旭东  颜鹏
作者单位:哈尔滨工业大学航天学院,哈尔滨 150001
基金项目:国家自然科学基金(61973101);;航空科学基金(20180577005);
摘    要:针对未知动态环境下异构无人机集群的分布式联盟形成问题,首先构建了异构无人机集群的联盟任务自动机,实现了异构无人机集群在未知动态环境下的自主任务协调.在此基础之上,提出了一种基于蒙特卡洛树搜索的分布式联盟形成算法,该算法通过两阶段的蒙特卡洛树搜索来增量式地对联盟结构进行分布式优化,算法可在分布式框架下实施,并在任意时间内...

关 键 词:异构无人机集群  分布式联盟形成  蒙特卡洛树搜索  分布式协同控制  未知环境
收稿时间:2021-09-03

A Distributed Coalition Formation Method of Heterogeneous UAV Swarm in Unknown Dynamic Environment
ZHENG Hong xing,GUO Ji feng,XIE Xu dong,YAN Peng.A Distributed Coalition Formation Method of Heterogeneous UAV Swarm in Unknown Dynamic Environment[J].Journal of Astronautics,2022,43(2):189-197.
Authors:ZHENG Hong xing  GUO Ji feng  XIE Xu dong  YAN Peng
Institution:School of Astronautics, Harbin Institute of Technology, Harbin 150001, China
Abstract:Aiming at the problem of distributed coalition formation of heterogeneous UAV swarm in unknown dynamic environment, we first construct the coalition task automata of heterogeneous UAV swarm to realize the autonomous task coordination of heterogeneous UAV swarm. On this basis, a distributed coalition formation algorithm based on Monte Carlo tree search is proposed, which increments the distributed optimization of coalition structure through a two stage Monte Carlo tree search. The algorithm can be implemented in a distributed framework and can terminate at any time to return the current optimal solution. Finally, the simulation results show that the proposed algorithm can effectively deal with the problem of distributed coalition formation of heterogeneous UAV swarm in an unknown and dynamic environment, and can maintain better performance in the large scale coalition formation problem.
Keywords:Heterogeneous UAV swarm  Distributed coalition formation  Monte Carlo tree search  Distributed cooperative control  Unknown environment  
点击此处可从《宇航学报》浏览原始摘要信息
点击此处可从《宇航学报》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号