Study Of Iterative Learning Control Based On Compositive Intelligent Algorithm

Iterative learning control theory is one important branch of the intelligent control theory. It was presented in the 80s of the 20th century. Iterative learning control algorithm is fit to control the repeated process, such as arm-and-hand system. Iterative learning control shows its benefits in the area of nonlinear system and other complicated system compare to other control algorithms. Furthermore, Iterative learning control algorithm is very simple and can be implemented easily.Study of iterative learning control is focus on the following item: convergence,robustness,initial value,learning speed and some others. As a new subject in control theory, study on iterative Learning control has got a lot of results, but there are more jobs still need to be done to extend the range of ILC applications.In this article, we introduce the development of ILC first. Depict the main content of the ILC study, after which we introduce the status of other intelligent algorithm study. In the 3rd part of the article, we present the sufficient conditions of convergence of the algorithm based on P type, D type, and PID type ILC. We compare the P type and D type ILC in converging velocity and Anti-interference capability. Consider that the D type ILC algorithm is sensitive to measurement noise; we provide a method to deal with the measurement noise. Simulation (based on Matlab 6.5) shows that we get good result in this problem. The practicality of D type ILC is improved. In the 4th part of this article, we do research on the integration of ILC and other intelligent algorithm, and provide a method to design a type of ILC controller based fuzzy gain turning. Simulation (based on Matlab 6.5) shows that this new control algorithm has advantage in increasing the converging velocity.