High performance motion controller design for linear piezoelectric ceramic motors

Abstract

This article presents a hybrid intelligent control and repetitive control design strategy for high performance motion control of linear piezoelectric ceramic motors. To compensate for the friction and hysteresis phenomenon in the piezoelectric motors, a radial basis function based neural network controller with PID controller is first applied as performance baseline for tracking periodic motion profiles. To further improve the periodic tracking performance, a discrete time repetitive controller coming from solving a particular solution of Bezout identity is then added to the feedback control system. The resultant hybrid controller consisting of neural controller, PID controller, and repetitive controller, is useful for tracking and disturbance rejection in motion control systems. The effectiveness of the proposed control strategy was verified through comparison of several tracking experiments on a linear piezoelectric ceramic motor.