An auxiliary model based on a recursive least-squares parameter estimation algorithm for non-uniformly sampled multirate systems

Abstract The lifted state-space models for a class of multirate systems non-uniformly sampled from their continuous-time systems are derived, and the corresponding input-output relationship is obtained. For unmeasurable information vectors arising in the identification models, this paper gives an au...

Full description

Saved in:
Bibliographic Details
Published in:Proceedings of the Institution of Mechanical Engineers. Part I, Journal of systems and control engineering Vol. 223; no. 4; pp. 445 - 454
Main Authors: Liu, Yanjun, Xie, Li, Ding, Feng
Format: Journal Article
Language:English
Published: London, England SAGE Publications 01.06.2009
SAGE PUBLICATIONS, INC
Subjects:
Algorithms
Computer simulation
Convergence
Data systems
Economic models
Estimating techniques
Excitation
Least squares method
Mechanical engineering
Monte Carlo simulation
Noise (mathematics)
Noise prediction
Parameter estimation
Parameter identification
Recursive algorithms
Sampled data systems
State space models
Variance
multirate systems
auxiliary model method
parameter estimation
recursive identification
state-space models
ISSN:0959-6518, 2041-3041
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Abstract The lifted state-space models for a class of multirate systems non-uniformly sampled from their continuous-time systems are derived, and the corresponding input-output relationship is obtained. For unmeasurable information vectors arising in the identification models, this paper gives an auxiliary-model-based recursive least-squares (AM-RLS) identification algorithm to estimate the parameters of non-uniformly sampled data systems using the auxiliary model method. The basic idea is to replace the unknown inner variables in the information vector with the outputs of the auxiliary model. Convergence properties of the algorithm proposed show that the parameter estimation error consistently converges to zero under the generalized excitation condition and bounded noise variance. A simulation example is included.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ObjectType-Feature-1
ObjectType-Article-2
content type line 23
ISSN:0959-6518
2041-3041
DOI:10.1243/09596518JSCE686