提出一种利用机器人完成空间桁架在轨装配的设计方案,并针对其中机器人的无碰运动规划问题,提出一种引入启发信息的运动规划方法.首先,设计了装配机器人和空间桁架系统;随后,引入基于距离场的碰撞检测算法进行空间中的障碍检测,并利用随机轨迹优化框架下的运动规划(stochastic trajectory optimization for motion planning,STOMP)算法进行机器人无碰运动规划;针对其中随机叠加轨迹方向不确定导致的规划效率低的问题提出了启发式STOMP(heuristically STOMP,hSTOMP)运动规划算法.仿真结果表明该算法能够快速得到机器人的无碰平滑关节轨迹,其中启发信息的引入大大提高了原算法的规划效率,具有一定的理论意义和应用价值.
Robot motion planning method for space truss on-orbit assembly
In this paper,a design scheme of using a robot to complete the on-orbit assembly of a space truss is proposed,and a motion planning method that introduces heuristic information is proposed for the collision-free motion planning problem of the robot.Firstly,an assembly robot and a space truss system are designed.Then,a collision detection algorithm based on distance field is introduced to detect obstacles in space,and the stochastic trajectory optimization for motion planning(STOMP)algorithm is used to plan the robot's collision-free motion.The heuristically STOMP(hSTOMP)motion planning algorithm is proposed to solve the problem of low efficiency caused by uncertain of the direction of stochastically-stacked trajectory.The simulation results show that the algorithm can quickly obtain the collision-free smooth joint trajectory of the robot.The introduction of heuristic information greatly improves the planning efficiency of the original algorithm,which has certain theoretical significance and application value.
万方数据
维普
