A New Iterative Learning Control Algorithm for Output Tracking of Nonlinear Systems

JL Kang,WS Tang,YY Mao
DOI: https://doi.org/10.1109/icmlc.2005.1527133
2005-01-01
Abstract:A new iterative learning control algorithm for nonlinear systems is presented. Sufficient condition for the convergence of this algorithm is given and proved. The new algorithm makes use of the exchange column updating method to construct the approximation of derivatives of the output function. The calculation work of the algorithm is largely reduced since the inverse matrix of the approximation of derivatives has the simple recurrence formula. Moreover, it is easy to show that convergence rate of the new iterative learning control algorithm is faster than that of the linear-type control updating law.
What problem does this paper attempt to address?