Calibration of FANUC Robot Dynamics Parameters based on the DOG LEG Method

Industrial robots are widely used in industrial manufacturing as an effective mean to improve working conditions, productivity. In recent years, machine is developed towards high speed and high precision, the static characteristics and the dynamic characteristics of industrial robots have become more and more demanding. The Newton-Euler method is used to establish the robot dynamics model in this paper. A calibration method of FANUC robot dynamics parameters based on the DOG LEG method is proposed, which includes: 1) a 6 parameters friction model was developed based on the Stribeck friction parameter, and a 60 parameters dynamics model was developed based on the inertia parameter, 2) we design the Fourier level excitation trajectory, and then operated the robot, following this trajectory, 3) we identified the above parameters by the DOG LEG algorithm. This model does not require mechanical sensors to be placed at the end of the robot and each joint. The experimental is implemented on the six degree of freedom FANUC robot. The results show that after calibrating the inertia parameter and friction parameter, the mean total joint moment error is reduced from around 52.78% to around 21.63%. The estimated dynamic model shows a significant improvement compared to the original model.