Dynamic Parameter Identification and Collision Detection of Robot

A robot dynamic parameter identification method based on improved linear friction is proposed. In this paper, the Newton Euler method is used to model the dynamics of the robot, and the dynamic expression is derived. Taylor expansion is used to linearize the commonly used nonlinear Stribeck friction model. Then the linear model and the minimum inertia parameter set of the robot are analyzed, and a set of minimum inertia parameters of the robot is derived. The linear expression of joint torque with respect to a set of identifiable parameters is obtained. The finite Fourier series is studied as the excitation trajectory of the dynamic parameter identification experiment. Combined with the minimum condition number, the optimization objective function of the excitation trajectory parameters is proposed, and the excitation trajectory parameters are calculated by MATLAB fmincon function. Fourier series is used as the excitation trajectory for experiment, and the identifiable dynamic parameter set is identified by the least square method. Using the dynamic parameters identified by the proposed method for dynamic modeling, the theoretical joint torque of the robot can be obtained, and the correctness and feasibility of the algorithm are verified by collision detection experiments.