A finite-time dead-zone modification for neural network iterative learning control

This paper presents the design of an iterative learning controller for a class of uncertain time-varying nonlinear systems in the presence of initial state errors. Through the introduction of initial rectified attractors and a finite-time dead-zone, a neural network iterative learning controller is designed with the proposed learning mechanism for the time-varying neural network training. The complete tracking is shown to be able to be realized over a pre-specified time-interval. In particular, all the signals in the closed-loop are guaranteed to be bounded, while the error between the system state and the desired one is ensured to converge to zero within a pre-specified time horizon. Moreover, the error is always confined to the dead-zone, which is well-governed by the introduced initial rectified attractor. Numeric simulation is carried out to demonstrate effectiveness of the proposed method.