Abstract: Aiming at the deviation between desired trajectory and ideal trajectory caused by the mechanism errors, and the dynamic error caused by the disturbance factors such as nonlinear friction and load change, which exist simultaneously in trajectory planning and trajectory tracking process of parallel robot, an error comprehensive compensation method for parallel robot is proposed. In the trajectory planning process, the pose error compensation is transformed to the driven rod parameters optimization problems based on the pose error model of parallel robot, then the particle swarm optimization algorithm is used to optimize driven rod parameters to correct the desired trajectory of parallel robot; In the trajectory tracking process, the dynamic error compensation strategy based on the adaptive iterative learning control algorithm is designed which can effectively track the desired trajectory. Simulation experiment on the Stewart platform is realized based on ADAMS and Matlab, which shows that the deviation of desired trajectory is corrected and the dynamic error of trajectory tracking can be compensated to realize the error comprehensive compensation of parallel robot. Furthermore, the workpiece machining experiment on the hybrid kinematics machine tools is done to test the validity of the method for improving the work accuracy of parallel robot.

Key words: Adaptive iterative learning control, Desired trajectory correcting, Dynamic error compensation, Parallel robot, Particle swarm optimization