-
TITLE: Multi-model system parameter estimation
-
AUTHORS: Ivan Markovsky, Sabine Van Huffel, and Bart De Moor
-
ABSTRACT: We pose a multi-model system parameter estimation problem. A multi-model system is a linearly parameterized system $H(z,p) = \sum_{i=1}^{n_p} p_i H_i(z)$. The parameter estimation problem is: given the set of systems $\{H_i(z)\}_{i=1}^m$, describing the multi-model system, find a causal system that assumes as an input the input/output signals of the multi-model system and produces as an output the parameter estimate.
We propose an easy to implement suboptimal solution. The algorithm that realizes it selects the best linear combination of the estimates produced by the Kalman filters designed for the models $\{H_i(z)\}_{i=1}^m$. ``Best'' is defined in the sense of minimization of the output error of estimation covariance.
The algorithm is appropriate for fault detection and can be viewed as an observer for the discrete state of a hybrid system.
-
STATUS: Published in the proceedings of the IEEE International Conference on Systems, Man, and Cybernetics, Hammamet, Tunisia, 6-9 October 2002. IEEE Catalog number 02CH37349C, ISBN: 0-7803-7438-X
- NOTE: The paper was nominated a finalist for the student best paper competition.
-
DATE OF ENTRY: 28 February 2001
@InProceedings{MVDM02,
author = {I. Markovsky and S. {Van Huffel} and B. {De Moor}},
title = {Multi model system parameter estimation},
year = {2002},
booktitle = {CD-ROM proceedings of the IEEE International Conference on Systems, Man, and Cybernetics},
note = {Available at: {\tt ftp://ftp.esat.kuleuven.ac.be/pub/SISTA/markovsky/reports/02-40.pdf}}
}