Enlighten
Research publications by members of the University of Glasgow
home > services > Enlighten

On transient dynamics, off-equilibrium behaviour and identification in blended multiple model structures

Murray-Smith, R., Johansen, T.A., and Shorten, R. (1999) On transient dynamics, off-equilibrium behaviour and identification in blended multiple model structures. In: 5th European Control Conference, 31 August - 3 September 1999, Karlsruhe, Germany.

[img]
Preview
Text
on2_transient_dynamics.pdf

548Kb

Abstract

The use of multiple-model techniques has been reported in a variety of control and signal processing applications. However, several theoretical analyses have recently appeared which outline fundamental limitations of these techniques in certain domains of application. In particular, the identifiability and interpretability of local linear model parameters in transient operating regimes is shown to be limited. Some modifications to the basic paradigm are suggested which overcome a number of problems. As an alternative to parametric identification of blended multiple model structures, nonparametric Gaussian process priors are suggested as a means of providing local models, and the results compared to a multiple-model approach in a Monte Carlo simulation on some simulated vehicle dynamics data.

Item Type:Conference Proceedings
Keywords:nonlinear system identification, multiple models, Gaussian Process priors, Bayesian statistics, nonparametric regression
Status:Published
Refereed:Yes
Glasgow Author(s):Murray-Smith, Prof Roderick
Authors: Murray-Smith, R., Johansen, T.A., and Shorten, R.
Subjects:T Technology > TK Electrical engineering. Electronics Nuclear engineering
College/School:College of Science and Engineering > School of Computing Science
Publisher:Springer
Copyright Holders:Copyright © 1999 Springer
First Published:First published in Advances in Control: Highlights of ECC '99
Publisher Policy:Reproduced in accordance with the copyright policy of the publisher

University Staff: Request a correction | Enlighten Editors: Update this record

Downloads per month over past year

View more statistics