Discrete-time Intermittent Iterative Learning Controller with Independent Data Dropouts

Abstract

In networked control systems (NCS), it is considered essential to design a robust controller such that the networked-system is stable against data dropouts during the network transfer. It has been shown that there is a critical data dropout rate over which the networked-system could be unstable; hence the desired task cannot be achieved. This paper shows that a desired task or trajectories can be still achieved even though there are feedback signal dropouts if the desired task is repetitive, as in the iterative learning control case. Specifically this paper shows how to design stochastic iterative learning control systems such that the networked-system with a repetitive task is robust stable against measurement and process noises and independent, intermittent output channel dropouts.