仿鹰鸽捕食逃逸行为的多无人机分组对抗博弈方法  

Multi-UAV grouping confrontation game method imitating the pursuit evasion behavior of eagles and pigeons

作　　者：仝秉达 段海滨[1,2] 魏晨 TONG Bing-da;DUAN Hai-bin;WEI Chen(School of Automation Science and Electrical Engineering,Bio-inspired Autonomous Flight Systems Research Group,Beihang University,Beijng 100083,China;Peng Cheng Laboratory,Shenzhen Guangdong 518055,China)

机构地区：[1]北京航空航天大学自动化科学与电气工程学院仿生自主飞行系统研究组,北京100083 [2]鹏城实验室,广东深圳518055

出　　处：《控制理论与应用》2024年第5期855-865,共11页Control Theory & Applications

基　　金：科技创新2030–“新一代人工智能”重大项目(2018AAA0100803);国家自然科学基金项目(U20B2071,91948204,T2121003)资助。

摘　　要：针对多无人机对抗问题,本文提出了一种三维空间中仿鹰鸽捕食逃逸行为的多无人机分组对抗博弈方法.在分析鹰鸽捕食逃逸行为的基础上,文章构建了多无人机博弈对抗系统模型,并定义了微分博弈中的连续可微值函数,证明了值函数满足Hamilton-Jacobi-Isaacs(HJI)方程,从而保证鞍点策略存在.使用最优分配方法,为仿鹰无人机一方设计了分组对抗分配策略,以解决多无人机追逃场景中的任务分配问题.本文对比仿真实验结果验证了所提出方法的有效性.In this paper,to solve the problem of multi-unmanned aerial vehicle(UAV)confrontation,a multi-UAV con-frontation game system is established and a multi-UAV grouping confrontation game method is proposed in three-dimen-sional space.The multi-UAV confrontation game system model and the continuous differentiable value function in the differential games are defined,and it is proved that the value function satisfies the Hamiltonian-Jacobi-Isaacs(HJI)equation ensurings that the saddle point strategy exists.For the problem of multi-UAV confrontation,a grouping assignment strategy is designed to solve the task assignment problem in the scenario.Comparative simulation results verified the effectiveness of the proposed method in solving the multi-UAV confrontation game problem.

关 键 词：多无人机对抗 微分博弈 任务分配 捕食逃逸 鞍点策略 

分 类 号：V279[航空宇航科学与技术—飞行器设计] TP18[自动化与计算机技术—控制理论与控制工程]