In order to improve the working accuracy of SCARA robots in industrial settings,it is one of the effective ways to analyze its dynamic model.The robot Lagrangian dynamic equation was established and linearized to obtain a set of linear expressions of the joint torque and parameters to be identified.The finite-term Fourier series was selected as the excitation trajectory model,and the joint angular velocity and angular acceleration were zero when the robot started and stopped to ensure the robot ran smoothly.Then the conditional number of the observation matrix in the linear expression was used as the optimization index,and combined with MATLAB optimization toolbox,the joint optimal excitation trajectory coefficient was obtained.Finally,the parameters to be identified were obtained by recursive least squares method,and substituted into the joint torque expression to compare it with the actual collected torque value to determine the change trend of the two.The results show that through experiment,a better effect can be obtained,and the joint torque calculated through the identification results can be used as a dynamic control in related fields.