Iterative learning control schemes for a class of nonlinear systems: Theory and real-time implementation

Abstract

Iterative Learning Control (ILC) is a powerful intelligent tool for stabilization of repetitive-task systems. However, choosing and fixing the gains of ILC schemes is not an easy task due to the fact that information propagation is present in two independent directions. In this paper, we propose new sufficient linear matrix inequality conditions for the design of iterative learning trackers for a special class of discretetime nonlinear systems. The control designs have been testified by the application of different real-time experiments using dsPIC 33 microcontroller and software in the loop simulation. The application of the developed results to speed control of a DC motor subject to a nonlinear friction shows satisfactory tracking performance in real time.

Adaptive PID-Type Iterative Learning Control for DC Motor Position

“…Iterative learning control (ILC) operates based on repeated tasks, which was proposed by Arimoto et al [12], updated law given in equation (11). Where ö, ÷, ù are learning gain matrices.…”

“…Finally, it is intelligent methods that overcome uncertainty problems well compared with others. Primarily, it can maintain high performance at a wide range of operations, for example, fuzzy control algorithm [6][7][8]; RBF neural network and Genetic algorithm [9]; and iterative learning control [10,11]. This paper is the first step in developing an autonomous steering system for a small self-develop electric car using a DC motor.…”