一种基于动态规划的并联机器人避障方法

Motion Intervenes Avoidance Algorithm for Parallel Manipulators With Dynamic Programming Method

为了解决并联机器人的运动干涉问题 ,将并联机器人的运动干涉统一为受限变量模型 ,提出一种基于动态调整指令轨线的运动干涉软防护策略。在运行过程的每一时刻 ,根据机器人运动学计算全部受限变量的当前值和速度 ;采用“虚拟制动过程”方法预测在考虑有限制动能力情况下 ,防止运动干涉的下一时刻可行速度 ,根据规划模型修改理想指令轨线在下一时刻的速度给定 ,给出了运动干涉动态软防护的算法。最后在研制的并联机器人的运行过程中进行了实验 ,结果表明 ,该算法有效地预测受限变量的临界状态 ,能使系统在不间断运行的情况下躲避干涉。

This paper deals with a motion planning strategy for parallel manipulator with dynamic constraints among intervenes.Given the values and velocities of intervene variables,according to the dynamic constraints,the admissible velocities of the next closely sufficiently state are efficiently computed.An algorithm is also presented for planning the optimal trajectory for a manipulator on a given path.Then,we apply the algorithm to an actual parallel manipulator,and evaluate its effectiveness by experiment. The result shows that it can detect the interference variable efficiently,and avoids the interference without intermission.