A Comparative Study of Feedforward Tuning Methods for Wafer Scanning Systems

Abstract

The use of feedforward control is beneficial for high-performance trajectory tracking in many motion control systems. Three methods of designing and tuning feedforward control signals (Iterative Learning Control, Iterative Controller Tuning, and Adaptive Feedforward Control) for a wafer scanner system are presented and compared. For this application, the main sources of tracking error are due to phase mismatch and nonlinear force ripple disturbance. The objective is to compare the performance of these three methods in compensating for error arising from these sources. The methods are compared based on a set of metrics. Comparison is followed by a discussion on advantages and disadvantages of each method including ability to reduce error during acceleration or scan phases of the trajectory, necessary assumptions, effect of inaccurate modeling, and effect of noise.