Instrumental-variable least-absolute-error algorithms for parameter estimation of continuous-time systems

Author

Kowalczuk, Z. ; Kozowski, J.

Author_Institution

Dept. of Autom. Control, Tech. Univ. of Gdansk, Gdañsk, Poland

fYear

1999

fDate

Aug. 31 1999-Sept. 3 1999

Firstpage

509

Lastpage

514

Abstract

The continuous-time (c-t) approach [good, saga, unbe1-2, youn] to identification of linear continuous-time models in the sense of least-absolute error (LA) and instrumental variable least-absolute error (IvLA) is under consideration. Estimates of the system parameters are obtained based on `finite-horizon´ formulation of the regression vector, which contains consecutive multiple `integrals´ of the input and output signals. The derived continuous-time algorithms and their continuous-time regression vectors are discretised and exercised numerically.

Keywords

continuous time systems; linear systems; parameter estimation; regression analysis; vectors; IvLA; continuous-time algorithms; continuous-time regression vectors; continuous-time systems; finite-horizon formulation; instrumental variable least-absolute error; instrumental-variable least-absolute-error algorithms; linear continuous-time model identification; parameter estimation; Covariance matrices; Estimation; Finite impulse response filters; Instruments; Measurement uncertainty; Numerical models; Polynomials; Parameter estimation; continuous-time systems; instrumental variable; least-absolute error; simulation;

fLanguage

English

Publisher

ieee

Conference_Titel

Control Conference (ECC), 1999 European

Conference_Location

Karlsruhe

Print_ISBN

978-3-9524173-5-5

Type

conf

Filename

7099355