A New Architecture for Robust Model Reference Control

Abstract

A new controller configuration for Robust Model Reference Control is presented in this paper. The structure is based on a right coprime factorization of the plant and makes use of an observer-based feedback control scheme combined with a prefilter controller. It is designed first to guarantee stability robustness and some levels of performance in terms of disturbance rejection. This is done by solving a constrained H ∞ optimization problem using the right coprime factorization of the plant in an active way. Secondly, the prefilter controller is designed to guarantee robust open-loop processing of the reference commands. This is done by assuming a Reference Model and by solving a Model Matching Problem.

Coprime factorization based strong stabilizing controller design

“…The selected factorization is intended to be optimal regarding the mixed sensitivity objective in (19). The same optimum pole placement idea for the selected factorization was already used in [9], but in this previous work the controller topology was slightly different and the strong stabilization issue was not tackled. The restriction in the optimization problem (19) ensures the stability of the obtained controller, given by (14).…”

“…Then it is possible to apply a Youla-type parametrization to obtain the set of observers that provide the partial state ξ. The set defined by (9) characterizes all linear observers for ξ [4]. Considering such a parametrization, the input/output relations (8) are rewritten in a way such that Q does not affect the command response and so it can be used to improve the robustness properties.…”

A Model Reference Based 2-DOF Robust Observer-Controller Design Methodology

“…In this chapter we present a 2-DOF control configuration based on a right coprime factorization of the plant. The presented approach, similar to that in (Pedret C. et al, 2005), is not based on setting the two Youla parameters arbitrarily, with internal stability being the only restriction. Instead, 1.…”