Repetitive variable structure control of micro-actuators with periodic disturbance and parametric uncertainties

The demand for higher accuracy and bandwidth motivates a dual-stage actuator system which circumvents the limitations of a single-stage positioning system. In this system, the micro-actuator arts as a fine positioning mechanism while the arm continues to perform coarse positioning. We developed a new repetitive learning system for micro-actuator which synthesizes variable structure control (VSC) and repetitive learning to improve tracking accuracy. Proof shows that the tracking error sequence caused by bias force, repeatable runout and uncertain parameters converges to zero on an extended L/sub 2/ norm. Repetitive learning in frequency-domain avoids the influence from high-frequency noises, resonance modes and other nonrepeatable components.