Dynamic Parameter Identification Of Industrial Robot Based On The Optimal Exciting Trajectory

Since the 21st century,robot research and industry development at home and abroad has been very rapid.The fundamental reason is that robots have the characteristics of efficient labor,which meets the needs of social production.On the other hand,the advancement of electronic science and control science,it also promotes the rapid development of robotics.In recent years,the continuous maturity of robotics technology has also prompted robots to be used in more high-speed,high-precision production applications.With the continuous development of robot technology,its integration with high-tech such as artificial intelligence and intelligent manufacturing is increasingly tight,and it also puts higher requirements on the control performance of the robot;Among the factors affecting the motion accuracy of the robot,in addition to the controller performance and other factors,the accuracy of the robot motion is largely determined by the accuracy of the dynamic model of the robot.Establishing an accurate dynamic model and adding it into the motion control of the robot can accelerate the convergence of motion error and improve its response characteristics.Therefore,an in-depth study of the dynamics model of the robot is needed.In this paper,the development status of robots at home and abroad is briefly introduced,which leads to the research objectives and technical problems.In this paper,the six degree of freedom of Efort ER3A-C60 industrial robot is taken as the research object,and the dynamic model is established according to the Newton-Euler method.The six degree of freedom of Efort ER3A-C60 industrial robot dynamics model is built.In particular,considering the situation of robot high speed motion,upon the foundation of the traditional joint torque plus friction torque model,the moment of inertia of motor and reducer are added to the robot torque model,which effectively reduces the joint torque estimation error.Then,the recursive robot dynamics model is linearized,and the minimum parameter set of ER3A-C60 robot dynamic parameter identification is obtained.The finite tract Fourier series is selected as the parameter to identify the excitation trajectory.The excitation trajectory parameters are optimized.In this paper,the linear equivalence constraint and the nonlinear inequality constraint are used to obtain the optimal value problem.In this paper,a genetic algorithm based on adaptive penalty function is designed to obtain the excitation trajectory optimization parameters.Finally,the identification experiment is carried out on the ER3A-C60 robot,and a dynamic identification method for the robot is proposed.Finally,the weighted least squares method is used to estimate the parameters of the robot dynamics parameters.According to the model,the experimental results show that the dynamic model established by the method is correct,and the dynamic parameters obtained in this paper have a good predictive effect on the joint torque of ER3A-C60 robot.