Control of Ultrasonic Motor by the NN-PI Controller using Variable Frequency

Abstract

In this paper, we propose PI controller combined with neural network (NN) for position and tracking control of ultrasonic motor (USM). In the conventional fixed gain PID controller, it is difficult to achieve precise control. Because the input-and-output characteristics of USM change with temperature rise or fluctuation of load-mass. In this design method, NN controller is used to compensate these characteristics fluctuation. Weights of NN are adjusted by back propagation method so that control error may be minimized. Moreover, variable frequency method is used for the control input in order to simplify drive circuit and realize quick response. The effectiveness of the proposed design method is confirmed by experiments using the existent USM.