Simulation Research On Inertia Feed-forward Control Of Flexible Joint Collaborative Robot Based On Online Identification

In this paper,the lightweight modular collaborative robot with flexible joints developed by the laboratory is taken as the research object.In order to solve the problem that the robot has inertial effect caused by the three joint modules at the end under high-speed,the theory of feed-forward control of the robot has been researched in depth.Different from common torque feed-forward control,this paper combines the double rigid body spring system and the online identification algorithm with the dynamic feed-forward,and proposes a new feed-forward control strategy with better real-time performance and more operability.This strategy can improve the tracking accuracy of the collaborative robot,speed up the response speed,improve the dynamic characteristics of the robot control system,and reduce the steady-state error of the robot under high-speed conditions.According to the structural characteristics and control requirements of the six-degree-of-freedom collaborative robots named "Bu Rui",the two rigid body spring system and the online identification algorithm are integrated into the feed-forward control theory to design a new feed-forward control system using dynamic model.Take the robot flexible joint as two rigid body spring systems,establish the block diagram model of the two rigid body spring system and derive its transfer function.According to the structure of collaborative robot,using improved D-H parameter method to establish the linkage coordinate system to get the robot kinematics model and the Jacobi matrix.Establish the dynamic model of the robot using Lagrange dynamic formula,and validate the correctness of the dynamic model by Solidworks,Adams and Matlab.Transform the transfer function of the two rigid body spring system into a difference equation by Z-transformation.Research the least squares identification algorithm in depth and establish the online identification algorithm with variable forgotten factor to identify the load inertia of the joint motor.Use the inverted hyperbolic tangent function to correct the value of the forgetting factor of the identification algorithm.Build the simulation verification platform of inertia online identification algorithm In the Simulink environment.Write The online identification algorithm using Matlab language in S-Function module.The identification algorithm is simulated according to two different operating states of the robot,and the result proves the correctness and superiority of the improved factor least squares identification algorithm.Build the simulation verification platform of the six-degree-of-freedom cooperative robot inertia feed-forward control technology In Simulink.Get the PID parameters of the robot servo control system by engineering method.Convert the three-dimensional physical model of the robot into positive dynamic model in SLX format by Adams,and take it as the controlled object of simulation.Make trajectory of the robot in joint space and cartesian space respectively,use it as the input of the simulation platform,give the quantitative index of trajectory tracking error.Compare the simulation results and the result proves the feasibility and advancement of the robot inertia feed-forward control technology proposed in this paper.