A High Precision Positioning Control Method Based on Learning Algorithm with Reference Model

Abstract

In this paper, a high precision positioning control method based on the learning algorithm with the reference model is proposed. The reference model is composed of a plant model and a feedback controller. In the proposed control method, disturbance, modeling error and nonlinear characteristic can be effectively compensated by the neural network-based controller, which learns the reference model. Moreover, the control-input saturation problem due to the over-learning for the neural network can be avoided. The effectiveness of the proposed control method is experimentally verified using the precision positioning equipment with nonlinear friction characteristics.