Research Of Manipulator Trajectory Tracking Control Based On Iterative Learning Algorithm

In recent years,domestic human resource costs have been rising,and domestic large-scale enterprises have introduced industrial robots instead of human to do some repetitive work that is complicated and heavy.However,these large enterprises often use robotic arms produced by some well-known foreign robot manufacturers.For SMEs,it's too expensive to purchase and maintain the robot arms produced by foreign manufacturers.So it is very practical to develop and produce our own industrial robot arms.Among the mechanical arms actually put into use in China,the number of four-axis robots ranked first,among which the planar joint type SCARA robot arm is used in a very wide range.In this paper,the SCARA manipulator is selected as the main research object,and the controller of which is mainly researched and designed.The iterative learning control algorithm is a control algorithm proposed in the1980 s.It is proposed in the background of robotic arm control.When the iterative learning control algorithm performs control,the structure and parameter information of the controlled object doesn't need to be known.It acquires various information about the controlled object by continually repetitive learning.ILC is very suitable for controlling the dynamic system with strong coupling,nonlinearity and various unknown disturbances.Aiming at SCARA robot arm,this paper proposes the corresponding control algorithm based on iterative learning control to improve the trajectory tracking control accuracy of the manipulator.The main work of this paper is as follows:(1)Thekinematics and dynamics analysis of SCARA robot arm is carried out,and its workspace and Lagrangian dynamic equation are obtained.And the motion trajectory planning was performed to obtain the desired target trajectory.(2)The basic principles,applicability conditions,learning laws and convergence analysis methods of the iterative learning control algorithm are studied and analyzed in detail.(3)The iterative learning control algorithm with variable gain is proposed.The MATLAB is used to simulate the learning control algorithm with variable gain and the general PD-type iterative learning control algorithm.The comparison of simulation experiments results proves the effectiveness and superiority of the iterative learning control algorithm with variable gain.In addition,simulation experiments of iterative learning control algorithm with variable gain are also carried out for the dynamic model of the robotic arm with different loads and uncertainties.The experimental results prove the robustness and adaptability of the proposed algorithm.(4)The adaptive iterative learning control algorithm is studied and proposed,and we proved its stability and convergence.The MATLAB is used to carry out the simulation experiments.The comparison with the experimental results of the iterative learning control with variable gain shows that the control accuracy of the adaptive iterative learning control algorithm is higher.The simulation experiments of adaptive iterative learning algorithm for manipulator dynamic models with different loads and uncertainties are also carried out.The experimental results proved the robustness and adaptive characteristics of adaptive iterative learning algorithm.The models of all simulation experiments in this paper are dynamic model of the SCARA robotic arm.The desired trajectories are selected in the workspace of SCARA robotic arm.