Non-Affine Nonlinear Stochastic Systems with Two ILC Update Laws under Random Data Dropouts

Abstract

This article for non-affine nonlinear stochastic networked systems investigates the convergence analysis and the tracking performance verification of two iterative learning control (ILC) update laws. In first ILC update law, if the information isn’t transferred, the update of the algorithm will stop. In second ILC update law, the update of the algorithm in each iteration will be continued using the newest accessible data even if no data is transferred in the current iteration. It is indicated that the input signals converge to the desired input, and no restrictive condition is imposed on the probabilities of the successful transfer of data. The convergence analysis of the two algorithms is based on concept almost sure. The comparisons of two presented ILC update laws are presented with a numerical example. Also, the tracking performance and effectiveness of the presented algorithms are shown.