Identification of robotic arm dynamics parameters based on semiparametric modeling

In order to solve the problem of low accuracy of parameter identification in the dynamics model of the robotic arm, a dynamics parameter identification method based on the semiparametric dynamics model is proposed on the basis of designing and optimizing the joint motion trajectory. Firstly, a connecting rod rigid body dynamics model is established based on the Newton–Euler method, and a linear regression recombination model is derived to obtain the minimum inertia parameter set and optimize the excitation trajectory by combining with a genetic algorithm; Secondly, the parametric rigid body dynamics model is identified based on the improved chaotic particle swarm algorithm, and a moment compensation model is established based on the bi-directional long and short-term memory network to compensate for and model the error of the non-parameterization The model of moment compensation is established; again, the data are preprocessed by combining IIR Butterworth low-pass filtering and RLOESS method, and the accuracy of dynamics identification is verified by combining with experimental trajectories; finally, the parameter identification test shows that the model has higher accuracy and robustness compared with the traditional dynamics model.