Ultrasonic motor control based on recurrent fuzzy neural network controller and general regression neural network controller

Abstract

The travelling-wave ultrasonic motor (TWUSM) has been used in industrial, medical, robotic and automotive applications. However, the TWUSM has the nonlinear characteristic and dead-zone problem which varies with many driving conditions. A novel control scheme, recurrent fuzzy neural network controller (RFNNC) and general regression neural network controller (GRNNC), for a TWUSM control is presented in this paper. The RFNNC provides real-time control such that the TWUSM output can tightly track the reference command. The adaptive updated RFNNC law is derived using Lyapunov theorem such that the system stability can be absolute. The GRNNC is appended to the RFNNC to compensate for the TWUSM dead-zone using a predefined set. The experimental results are shown to demonstrate the effectiveness of the proposed control scheme. © Springer-Verlag Berlin Heidelberg 2013.