Research on PD-type Iterative Learning Control Algorithm of Manipulator Based on Gain Switching

Multi-joint serial manipulator is highly nonlinear and strongly coupled system with multiple inputs and multiple outputs. In addition, there are some problems such as parameter perturbation, external interference and uncertainty are exist. The trajectory tracking control of manipulator has certain difficulties. As the most important part of manipulator system, the design of controller is related to the accuracy and robustness of manipulator to perform tasks. For hydraulically driven manipulators, it is reciprocate in working process, and has the characteristics of long delay time. Aiming at characteristics and problems of multi-joint series manipulator, this paper proposes a PD-type iterative learning control algorithm based on gain switching, it has the robust term, and the linearized residual is considered to realize the control of the mechanical system with uncertain dynamics equation and non-repetitive interference. The first two joints of the six degrees of freedom manipulator XIOPM, which developed by the research group are simulated by MATLAB/SIMULINK using this method. Compared with the traditional iterative learning control, the algorithm proposed in this paper can converge the actual output trajectory of the system to the desired output trajectory at a relatively fast speed with good tracking performance and high tracking accuracy.