|
RECURSIVE IDENTIFICATION OF LINEAR MULTIVARIABLE SYSTEMS WITH BOUNDED DISTURBANCES
M. Boutayeb‡, Y. Becis‡* and M. Darouach‡
‡ CRAN CNRS University of Henri Poincaré, IUT de Longwy
186, rue de Lorraine, 54400 Cosnes et Romain, FRANCE
* CRP-HT, LTI group, Technoport Schlassgoart. 66, rue de
Luxembourg BP 144 4002 Esch/Alzette, LUXEMBOURG.
e-mail: Yasmina.Becis@iut-longwy.uhp-nancy.fr
In this contribution a simple and useful recursive identification method for linear multivariable systems in the presence of bounded disturbances is presented. We show in particular how to design a selection rule for some weighting matrices so that the estimated parameters are consistent with the measurements and the noise constraints. Conditions for asymptotic convergence are given. Performances and simplicity of the design will be shown through a numerical example to estimate physical parameters of a mechanical system.
Keywords: Parameter identification, Multivariable systems, Recursive estimation, Bounded noise
Session slot T-Th-M21: Posters of Design Methods and Optimal Control/Area code 2b : Linear Systems
 | 