基于双层博弈的多臂在轨服务航天器路径规划

针对在轨服务多臂航天器系统高精度的位姿协同要求及其运动过程中的避障约束，提出一种基于机械臂末端（腕关节）和肘关节的双层博弈多臂路径规划方法。研究建立了多臂运动学模型，在博弈论基础上建立多臂的博弈模型；给出了双层博弈的基本算法流程及其纳什均衡解的求解策略；以动目标多臂围捕为场景进行仿真分析，验证所提出算法末端精确跟踪抓取和肘部避障能力的有效性和实用性。所得结果可为多臂在轨服务航天器的智能化路径规划与控制提供新的解决方案。 

Two-level Game Based Multi-arm On-orbit Servicing Spacecraft Path Planning

A two-level game strategy is proposed for the on-orbit servicing spacecraft multi-arm path planning and collision avoidance. The first-level game, i.e., end-effector game, is utilized to guarantee the capturing path planning of multi-manipulator while the second-level game, i.e., elbow joint game, is used to avoid self-collision during motion and govern a balanced distribution between multiple arms. The kinematics for the multi-manipulator is established and the game model of multi-manipulator is formulated on the basis of the game theory. Then the basic process of the proposed two-level game is given as well as the algorithm for obtaining the Nash equilibrium to solve the proposed two-level game strategy. Simultaneously, some simulation analyses are carried out under the scenario of multi-arm spacecraft round-up a moving target case to demonstrate the effectiveness and practicality of the proposed method. Simulation results verify the performance of the proposed two-level game in high precision end-effector planning and elbow joint collision avoidance in the capture process. The proposed method provides a new solution for intelligent path planning and control deployed multi-arm on-orbit servicing spacecraft.