Analysis And Design For Some Control Problems Of Robot System

Presently,the overall development level of robotic manipulator which has become the key technology in the field of intelligent manufacturing and industrial automation indicates the level of industrial modernization and the comprehensive national strength of a country.In this paper,the position tracking control of industrial robot based on joint space and image-based visual servo control are researched by means of adaptive control and iterative learning control.Those are carried out under the conditions of uncertain model and disturbance in the robot system,visual servo of image Jacobin matrix pseudo inverse in dynamic and so on.The main work of this paper is as follows:1.According to the characteristics of repeatability in robot's task and the presence of uncertain parameters and external disturbance in trajectory tracking,several control strategies of adaptive iterative learning with variable gain are presented.The controllers contain the part of PD feedback and the learning part used for parameters uncertainty and disturbance of robotic system.The stability and asymptotical convergence are proved by the Lyapunov-like function.The simulation verifies the feasibility of the algorithms.2.Aiming at the problems that there are many iterative parameters and the initial running point is limited in the iterative learning control of robot,the proposed learning controller applies the positive definite learning coefficient matrix to the trajectory tracking error and error derivative and the updated two-dimensional learning control quantity is obtained.Iteration parameters are reduced to two,which is of great significance for saving storage space and improving operation speed.Assuming some system parameters are bounded,the iterative learning parameters can be further reduced to 1.The proposed new control strategy can relax the initial condition that the system should be in the initial state of the desired position each time the system runs(the initial correction condition of the robot control system is relaxed),which means that the operation of the system starts at the end of the last operation,instead of starting from the same starting point every time.3.An adaptive robust iterative learning control method is proposed for a class of robot systems with uncertain dynamic models and non-repetitive disturbances.The new method combines variable gain PD and iterative learning feedback control,and has self-adaptability and learning ability in the case of simple control structure.By periodic iteration,the current control information is updated by the previous.The new method updates the control gain in a form similar to switching control,which increases the control flexibility and makes the trajectory tracking convergence speed of the robot system faster.The asymptotic convergence is verified by Lyapunov function.The simulation shows that the position error and speed error of the robot system will decrease monotonically as the number of iterations increases.4.Aiming at the problem that robot system is susceptible to external interference and internal parameter changes,two adaptive switching control strategies with good tracking performance,excellent control quality and robustness are proposed for the trajectory tracking problem of robotic manipulator in this paper.The first scheme is designed in the condition that the supremum of the bounded disturbance for robot manipulator is known;the second scheme is proposed when the supremum is unknown.Both of these two controllers include an adaptive switching law and a PD controller.Based on the Lyapunov stability theorem,it is shown that the proposed two new schemes can guarantee tracking performance of the robotic manipulator and be adapted to the alternating unknown loads.Simulations of two-link robotic manipulator show that the two schemes can avoid the overlarge input torque,and the feasibility and validity of the proposed control schemes are proved.5.In order to solve the problem that it is necessary and difficult to obtain the image Jacobian matrix(interaction matrix)or its inverse matrix in uncalibrated visual servoing for image-based dynamic control of robot manipulators,a new dynamic uncalibrated visual servo method which can obtain the image Jacobian pseudo inverse matrix is proposed.In this method,neither camera's intrinsic and extrinsic parameters nor joints' speed of robot is requisite.It is proved that the image error has asymptotic convergence by Lyapunov theory.Simulations have been conducted to verify performances of the proposed controller.Results demonstrate the good convergence of image errors.