Reduced-Order Multiple Observer Design for Takagi-Sugeno Models with Unknown Inputs

Abstract

The paper presents the design of a new reduced-order multiple observer (ROMO) for state estimation of unknown input Takagi-Sugeno (T-S) models; this is only the second reduced-order multiple observer ever designed for such systems. The proposed observer is obtained by applying the classical methods of reduced-order observers' design for linear multivariable systems with unknown inputs to the case of T-S models. The originality of the designed observer compared to the only existing one is related to: 1) no existence conditions are required, 2) no use of the pole placement method leading to sensitivity problems is necessary and, 3) the solution is designed using available commercial LMI tools. The related algorithm of the proposed ROMO is structured into five steps; its efficiency is proved by using a concrete example of a light aircraft lateral-directional motion.