基于改进DE算法的工业机器人时间最优轨迹规划

An Improved DE Algorithm for Time Optimal Trajectory Planning of Industrial Robot

为提高工业机器人的工作效率,提出一种进行时间最优轨迹规划的新算法。通过对已知任务轨迹的关键点进行运动学反解,求解与之对应的关节空间位置序列,并采用5次非均匀有理B样条曲线构造关节运动曲线,能够保证机器人各关节位置的准确性,实现各关节运动的速度、加速度以及二次加速度的连续性。通过改进差分进化(Differential Evolution简称DE)算法,充分利用不可行解的信息,加强对边界的搜索,增强了算法的全局搜索能力。与遗传算法以及差分进化算法进行比较,利用该算法进行轨迹规划,结果显示该算法的搜索速度更快,所得的数值结果更小。

In order to improve the efficiency of industrial robots,a newalgorithm for time optimal trajectory planning is proposed. Position series in joint space are obtained by applying inverse kinematic algorithm to a specified trajectory in task space. The 5-times non-uniform rational B-spline curve is used to construct the joint motion curve,which can ensure the accuracy of the joint position of the robot and also achieve continuity of speed,acceleration and secondary acceleration. By improving the differential evolution（DE） algorithm,infeasible information is fully utilized,strengthen the search of the boundary and enhance the global search ability of the algorithm. Comparing with the genetic algorithm and the differential evolution algorithm,the algorithm is used to carry out the trajectory planning. The result shows that the algorithm is faster and the result is smaller.