空间多机器人协同运动规划研究  

Research on Cooperative Motion Planning of Space Multi-robots

作　　者：关英姿[1] 刘文旭 焉宁[2] 宋春林 GUAN Yingzi;LIU Wenxu;YAN Ning;SONG Chunlin(Department of Astronautics Engineering, Harbin Institute of Technology, Harbin 150001;R & D Centre, China Academy of Launch Vehicle Technology, Beijing 100076)

机构地区：[1]哈尔滨工业大学航天工程系,哈尔滨150001 [2]中国运载火箭技术研究院研究发展中心,北京100076

出　　处：《机械工程学报》2019年第12期37-43,共7页Journal of Mechanical Engineering

基　　金：国家自然科学基金资助项目(91648201)

摘　　要：面向空间在轨装配任务提出基于优先级的多机器人协同运动规划方法。多机器人系统包括一个7自由度操作机器人和一个13自由度超冗余照明机器人。采用两种规划方法规划高优先级的操作机器人,即关节空间点到点的离线路径规划和基于伪逆的笛卡尔空间在线运动规划。针对低优先级的超冗余照明机器人,将操作机器人视为已知障碍,采用基于A^*算法和末端轨迹跟随的可重构运动规划方法,将运动规划分解为末端路径规划和机器人各关节对轨迹的跟随问题,并验证算法的可重构性。采用具有特定运动学形式的机器人模型进行仿真试验,仿真结果表明,提出的各个机器人的运动规划方法均能有效满足任务需求,且照明机器人的运动规划算法具有可重构性,多机器人系统基于优先级的协同策略能够满足空间在轨装配任务需求。A multi-robot cooperative motion planning method based on priority is proposed for space on-orbit assembly task. The multi-robot system is composed of a 7-DOF manipulator and a 13-DOF super redundant lighting robot. Two methods are used to plan high-priority manipulators: off-line path planning from point to point within configuration space and on-line motion planning in Cartesian space based on pseudo-inverse. For the low-priority super-redundant lighting robot, the manipulator is regarded as a known obstacle, and a reconfigurable motion planning method based on A^* algorithm and trajectory tracking is proposed. The motion planning is decomposed into path planning and tip-following, and the reconfiguration of the algorithm is verified. Simulation experiment is conducted with two robot kinematics models. Simulation results show that the motion planning method of each robot can meet the task requirements effectively, the motion planning algorithm of lighting robot is reconfigurable. The cooperative strategy based on priority of multi-robot system can meet the requirements of space on-orbit assembly task.

关 键 词：在轨装配 多机器人系统 协同 运动规划 超冗余机器人 

分 类 号：TG156[金属学及工艺—热处理]