Real-time optimization for nonlinear systems using algorithmic control

Authors:	Kobayashi Tomoaki, Osaka Prefecture University, Japan Imae Joe, Osaka Prefecture University, Japan Magono Michiyuki, Nara Institute of Science and Technology, Japan Yoshimizu Kazutaka, Osaka Prefecture University, Japan Zhai Guisheng, Osaka Prefecture University, Japan
Topic:	2.1 Control Design
Session:	Numerical Issues in Systems and Control
Keywords:	Real time optimization, nonlinear system, algorithmic design

Abstract

A new method for on-line optimal control of nonlinear systems is proposed. The proposed method is based on the algorithms for obtaining the numerical solutions of optimal control problems. According to this idea, we can use the iterative computational method for the on-line optimal control problems. In this paper, we adopt the Riccati-equation based algorithm, one of the iterative algorithms, and demonstrate the proposed algorithmic method is applicable to such mechanical systems as fast-sampling is required.