Research On Control Of Visual Feedback Arm

The traditional system model based on the manual arm control system has been difficult to deal with the uncertainty of the system parameters and reject the external interference,it is difficult to make th e tracking error of the robot arm quickly converge.The information obtained from the visual information is fed back to the control system to improve the tracking accuracy of the robot arm.The transformation of the target location and the coordinate syste m in the image processing is analyzed,which provides the basis for the expectation trajectory of the repeated control system.Based on the learning algorithm of visual feedback robot control algorithm,this paper adopts the iterative learning algorithm ba sed on gain matrix estimation to improve the tracking accuracy of the robot in the repetitive task.In this paper,the positive and inverse kinematics and dynamics of the manipulator are analyzed.By comparing the convergence speed of the trajectory tracking error of the robot arm simulation with different degrees of freedom and the different gain matrix iterative learning control algorithm,the tracking of the repeated trajectory Accuracy problem.Based on the radial basis function neural network iterative learning control optimization control law,the gain parameter is proposed for the PD type iterative learning control trajectory tracking error convergence rate is not ideal.The simulation results show that the optimized control algorithm improves the convergence speed of the trajectory tracking error of the manipulator.Aiming at the influence of different gain matrix on trajectory tracking control,the method of solving the frequency domain characteristic equation of the control system is adopted,and the gain matrix of the repeated controller is transformed into the tilt gain matrix to improve the error convergence rate of the system.The simulation results show that the control system based on the tilting matrix improves the convergence speed of the trajectory tracking error of the manipulator and provides a new method for tracking the trajectory tracking accuracy of the control system based on the convergence rate of the trajectory tracking error of the neural network iterative learning control.The act ual gain matrix data that improves the convergence rate of the error.In this paper,the error convergence performance of the tracking trajectory of the manipulator is improved by completing the optimized iterative learning controller design.On the basis of the above research,ROS operating system and V-REP software platform which can be seamlessly connected with the actual manipulator are simulated,and finally based on VS software platform.