Reduced-order Multiple Observer with Application to the State Estimation of a Controlled Generic Reentry

Abstract

The paper presents the design of a new reduced-order multiple observer to achieve the state finite-time reconstruction for Takagi-Sugeno multiple models with unknown inputs. The new observer is obtained by combining a classical reduced-order observer for linear multivariable systems with unknown inputs and a full-order multiple observer for Takagi-Sugeno systems. The new observer has no existence condition due to a “while” loop and to the used techniques (coordinate change and QR decomposition). The main innovative idea behind its obtaining is to divide the multiple model into two parts: an unknown-input-free subsystem and an unknown-input-depending subsystem. The new observer has been software implemented and validated for a reentry vehicle’s motion.