Research On Industrial Robot Inverse Dynamics Control And Parameter Identification Method

Industrial robots are an important foundation for intelligent manufacturing and are widely used in automotive,electronics and light industry production lines.The control system is the soul core of the industrial robot,and its performance determines the processing quality and operating efficiency.Since the robot is an articulation system,the motion inertia changes with the movement posture and the joint torque coupling degree is high.When moving in high speed,heavy load and arm span change greatly,the inertia changes greatly,and the nonlinear effect is significant.When using conventional PID control methods,only conservative control parameters can be used to ensure stability,which results in poor dynamic characteristics,large trajectory errors and low positioning efficiency,and system performance cannot be exerted.To this end,this paper studies industrial robot control technology based on inverse dynamic feedforward to improve robot control performance and operating efficiency.The main contents are as follows:The 6-DOF industrial robot was dynamically modeled and analyzed,and the joint friction torque was linearized;the feedforward control strategy based on inverse dynamics was studied,and the control performance simulation was carried out in Simulink environment.The results show that the inverse dynamic feedforward control can effectively improve the robot's dynamic characteristics and control performance.The key to inverse dynamics control lies in the accurate acquisition of model parameters.To this end,the overall identification method is used to identify the robot dynamics parameters: the dynamic model is simplified with the minimum inertia parameters;the excitation trajectory designed by the finite Fourier series + quintic polynomial method,The objective is to optimize the trajectory parameters by minimizing the condition number of the observation matrix,and to improve the robustness of the excitation trajectory to noise;use the weighted least square method to estimate the parameters;and use the extended state observer to process the joint motion sampling signal.The simulation results show that the method Superiority.On the basis of the above research,a 6-DOF industrial robot experimental platform was built,and the robot dynamic parameter identification and motion control experimental research based on inverse dynamic feedforward were carried out.The results show that the parameter identification method achieves a good identification effect,and the feedforward control based on inverse dynamics has obvious effect,which can effectively improve the robot control characteristics and effectively reduce the trajectory error.