机器人模糊迭代学习控制及其仿真研究（Fuzzy iterative learning control of robot and Its Simulation Research）.docVIP

机器人模糊迭代学习控制及其仿真研究（Fuzzy iterative learning control of robot and Its Simulation Research） Fuzzy iterative learning control of robot and Its Simulation Research Abstract: in the iterative learning control for trajectory tracking of robot manipulators, iterative learning law is difficult to choose, combining with the basic idea of self correcting control, fuzzy logic and iterative learning control, adaptive learning rate method to determine the fuzzy control, and using the Matlab software of the Simulink method is applied to the trajectory of the robot with high precision the tracking control is simulated, the results show that this method has the advantages of simple and practical learning control law, high tracking accuracy, fast learning speed and strong robustness, etc.. Keywords: iterative learning control fuzzy control simulation research in robot control 1 Introduction Trajectory tracking control of robot is an important kind of control in robot control, and many control methods have been proposed. The iterative learning control (Iterative Learning Control, ILC) by using iterative method, can complete tracking ideal input arbitrarily complex, is considered to be an effective solution, and in order to an important branch of intelligent control, it is the first by the Japanese scholars in the mountain (M.Uchigama) is proposed to control problem in 1978 Study on high speed movement of the manipulator 1. Over the past two decades, the research of iterative learning control has developed greatly, and some achievements have been made. [2-8] in the iterative learning control research, the learning rate is the key. In recent years, Arimoto and his collaborators first proposed D type learning law, 9 - 10, after they have made the PD type and PID type learning law; Craig first proposed filter type learning law 11; application of historical data of the southeast iteration, Wu higher order learning law 12; Heinzinger et al. D type learning law with forgetting factor 13; Ren