Dynamic identification of Staubli RX-60 robot using PSO and LS methods

摘要

This paper deals with the dynamic modeling and identification of Staubli RX-60 robot. In the robot identification, a least squares (LS) method and particle swarm optimization (PSO) technique were used to estimate the distinct inertia parameters of Staubli RX-60 robot. Several experiments were conducted to have the physical robot data. In identification experiments, the position, velocity, acceleration and torques of the robot joints were measured from the motor encoders, motion analysis system with three cameras and the load-cell sensor. Using experimental data, the inertial parameters of the robot were successfully estimated. The parameters estimated from these methods were verified with experimental results. These experimental results show that the estimated inertial parameters predict robot dynamics well. Moreover, the identification errors for both PSO based identification technique and LS method were computed and were summarized in a table. According to the identification errors, the performance of PSO on the parameter estimation is better than the performance of LS method.