Soichi Ibaraki, Kyoto University

This paper presents an algorithm to solve an H-infinity optimization problem of fixed-structure controllers. The proposed algorithm first transforms the original fixed-structure controller synthesis problem into an H-infinity synthesis problem of static output feedback control, which does not impose any structural constraints except for the one on the controller order. Then, a linearization-based local search algorithm is applied to solve this problem. Based on the proposed H-infinity optimization algorithm, this paper demonstrates an intuitive and efficient method to re-tune controller parameters, which were finely tuned by an expert engineer, and improve the control performance. As an application example, the proposed approach is applied to the tuning of a PID (Proportional plus Integral plus Derivative) controller for head positioning of hard disk drives. The effectiveness of the proposed re-tuning method is demonstrated by simulation and experimentation.