Skip to Main Content
A wavelet neural network (WNN) based two-axis motion control system using ultrasonic motor (USM) drives is proposed. First, a driving circuit for the USM, which is composed of a push-pull DC-DC power converter using PWM direct duty-cycle control and a current-source two-phase parallel-resonant capacitor-parallel load inverter, is introduced briefly. Moreover, since the dynamic characteristics of the proposed two-axis motion control system are complicated, an online trained WNN control scheme with varied learning rates is proposed to control the position of the two-axis moving table with high-performance features. In addition, a reference-word circular interpolator technique, which can permit the simultaneous operation of two-axis motion, is described. The effectiveness of the proposed motion control strategy is demonstrated by some experimental results.