Novel derivation for the polynomial optimal control methods. 2. GLQG controller design

Abstract

This paper presents a novel derivation for the problem of polynomial generalized-linear-quadratic-gaussian (GLQG) control following a systematic approach for the derivation and considering a more general plant-structure that contains colored input disturbance and measurement noise. The presentation of the theory comes in a more concise, clear and general form to help those looking to use it without any details as well as those looking for detailed understanding and tailoring the theory to their problems. The cost function includes dynamic weighting elements allowing integral action to be introduced and robustness characteristics to be modified. Thus, the novelty of the paper stems from the fact that it presents the proof in a novel approach for a general plant structure which covers and could be broken down easily to any special case that might exist in reality. The paper is supplemented with design steps and two numerical examples: one is a continuous time system and the other is a discrete time system.