The Design Of Iterative Learning Algorithm And Study On Its Convergencr Speed

The purpose of Iterative learning controller is repeating operation again and again to achieve high precision fast tracking expected trajectory.The beginning of the article analyses several main iterative learning control law and its development,Scholars have preliminary knowledge of iterative learning control.Although the iterative learning controller can compensate for unknown repetitive error and interference,its own is not enough,it needs to combine with other methods.Iterative learning algorithm is widely used,it can be applied to the mobile robot,the robot's mechanical arm,industrial robots and other aspects.Based on iterative learning control algorithm in robot trajectory tracking research,the main research content is as follows:First of all,we propose two new iterative learning control algorithm,with the state and the output measurement noise exist disturbance mobile robot system.It can improve the convergence speed and enhance the system robustness and anti-interference.The stability of the system is proved in theory,it guarantees the approximate zero tracking error convergence.The simulation results verify the proposed algorithm.Secondly,we first put forward the higher order feedback-feedforward iterative learning algorithm for a class of uncertain items and the repeatability of nonlinear system interference.Then we design the improved iterative learning control system to track the desired trajectory,its algorithm can advance compensation by disturbance weaken the influence of the output tracking,strict mathematical deduction and simulation results show that the controller can efficiently achieve the perfect tracking.Last,in iterative algorithm applied in practical engineering,with study the convergence rate of the iterative learning algorithm,the convergence speed has been one of main indexes and the judgment criteria for iterative algorithm.Here are several judgment methods of convergence speed of the iterative learning control algorithm to compare the different convergence speed of iterative learning control provides the basis and the reference standard.