Robust Periodical Tracking for Fast Tool Servo Systems With Selective Disturbance Compensation

Abstract

Fast tool servo (FTS) systems have emerged as a promising technology in ultraprecision manufacturing, where sophisticated contouring in the manufacturing process poses significant challenges to the dynamical tracking of periodical references at nanoscale. The situation is further complicated by the existence of time delays, load variations, as well as disturbances. Comprehensive handling of these problems in a unified framework is crucial to the manufacturing precision of FTS systems. In this paper, we propose a modified robust repetitive control structure for FTS systems with time delays to improve the tracking performance by loop shaping. On top of the specific structure of the robust repetitive controller, a parallel structure is first introduced to optimize the low-pass filter to increase controller gains at the fundamental and harmonic frequencies, such that an improved tracking performance is obtained but at the cost disturbance rejection capabilities. To this end, a cascaded structure is then introduced to further shape the sensitivity function to compensate the external disturbances especially near the maximum sensitivity frequency. Taking the robustness against model uncertainties into consideration, we further formulate the multistage design of the proposed controller as an H∞ optimization problem solving by skew Toeplitz approach. Real-time experiments are conducted on a prototype of the FTS system demonstrating excellent tracking performance of the proposed control approach.