A learning approach to precision speed control of servomotors and its application to a VCR

This paper describes a learning control method that gives a low-cost solution to the high-precision speed regulation problem for servomotors in the presence of unknown and unstructured disturbance torque. The highlight of the learning control method is a "functional" update strategy based on steady-state oscillations of speed error. To show the generality of our work, we mathematically characterize the steady-state properties of the proposed learning system and then give a rigorous analysis for the convergence of the proposed update scheme. Furthermore, our experimental results using a commercial home-use VCR show that the proposed learning control method can be implemented at a low cost but can remarkably reduce the speed error.