Chattering-free adaptive iterative learning control for linear-motor-driven gantry stage with initial state errors

This paper aims to investigate a chattering-free adaptive iterative learning strategy for trajectory tracking tasks on linear-motor-driven-gantry-stages with initial state errors. The reduction mechanism of the proposed parameter learning law effectively addresses initial state tracking errors, unmodeled system dynamics, and external disturbances during the iterative process. Furthermore, to mitigate the oscillation issues caused by traditional iterative learning control signals, this paper adopts the approximation for the sign function. However, this approximation introduces the non-negative definite problems. Therefore, this paper introduces a novel analysis method based on contraction mapping and composite energy functions in the Lyapunov-like theory. This method rigorously proves the boundedness and convergence during the entire iteration under non-negative definite problems with initial state errors. The effectiveness of the proposed approach is validated through experiments on a linear motor platform.