Autonomous Parameter Design for Cascade-Structure Feedback Controller Based on Cooperative Optimization Method

Abstract

This paper presents a cooperative optimization-based autonomous parameter design method for cascade-structure feedback (FB) controllers used in industrial precision servo systems. Designing fine parameters for a cascade-structure FB controller that realizes a wide control bandwidth is a nonconvex optimization problem owing to the nonlinear relations between the parameters. In conventional parameter design methods based on nonlinear optimization techniques, the obtained parameters strongly depend on the initial point, which requires considerable labor for trial and error-based initial point setting in industry. This paper aims to eliminate such annoyance in cascade-structure FB controller design by using the cooperative optimization technique with a simple initial parameter setting method. The effectiveness of the proposed method is verified through an example FB controller design for a galvanometer scanner, compared to a sequential quadratic programming-based parameter optimization method.