Identification Method of Robot Dynamics Parameters Based on Improved Friction Model

Abstract

Follow-up bracket mechanism has been widely used in airborne equipment as a special robot. Due to the high complexity of the mechanism itself and the need to meet the continuous time variability of the tracking trajectory, the dynamic analysis of the Follow-up bracket mechanism must be carried out. In this paper, the MDH method is used to analyze the necessary kinematic parameters of the Follow-up bracket mechanism, and the dynamic model is constructed by the Newton-Euler iteration method, and the dynamic expression is deduced by the symbolic calculation software SYMORO+; then the linearization model of the robot and the minimum inertial parameter set is used to derive a set of minimum inertial parameters for the six-degree-of-freedom industrial robot; the finite-term Fourier series is studied as the excitation trajectory of the dynamic parameter identification experiment, and using MATLAB fmincon function in the calculation of the excitation trajectory parameters; research on the least squares parameter identification method. And it is proposed to modify the traditional linear friction force identification model into a sigmoid function model to avoid the problem of large error at joint commutation and control law chattering. Experiments show that the theoretical value of joint torque calculated by the deduced linear expression is consistent with the change trend of the actual value. The dynamic parameter identification method based on the improved friction model can optimize the theoretical joint torque when the interactive mechanism moves, and improve the chattering problem at the joint reversal.