基于改进粒子群算法的并联机器人运动学精度提高新方法

针对并联机器人在基于给定工作任务进行轨迹规划过程中,存在因机构误差引起的期望轨迹与理想轨迹之间的偏差,由此造成并联机器人运动学精度降低的问题,提出了一种并联机器人运动学精度提高新方法。首先将连续工作任务离散化为满足精度要求的若干理想位姿点,在建立并联机器人位姿误差模型基础上,将机构误差项转化为驱动杆误差;基于种群排列熵模型和粒子速度激活机制改进了粒子群算法,并利用改进的粒子群算法组合优化驱动杆参数,补偿并联机器人位姿误差,进而修正期望轨迹以提高并联机器人运动学精度。通过MATLAB和ADAMS仿真验证了所提出方法的可行性和有效性。

A New Method for Improving Kinematics Accuracy of Parallel Robot Based on Improved PSO

Aiming at a kinematics accuracy problem of the parallel robot,caused by the deviation between the desired trajectory and the ideal trajectory generated by the mechanism errors based on the task trajectory planning process,a new method to improve the kinematics accuracy of the parallel robot was proposed herein:First,the continuous task was discretized to some ideal pose points based on the accuracy requirements,and the mechanism errors were transformed into drive rod errors based on the pose error model of the parallel robot.Then an improved particle swarm algorithm was developed based on the population permutation entropy and the particle velocity activation mechanism,which can complete combinatorial optimization of the drive rod parameters for compensating the pose errors of the parallel robot,so as to correct the desired trajectory and improve the kinematics accuracy.The feasibility and validity of the proposed algorithm were shown by the MATLAB and ADAMS simulation.