On-line Nonparametric Regression to learn State-Dependent Disturbances

Bas J. de Kruif; Theo J.A. de Vries

doi:10.1109/ISIC.2003.1253917

On-line Nonparametric Regression to learn State-Dependent Disturbances

Bas J. de Kruif, Theo J.A. de Vries

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

49 Downloads (Pure)

Abstract

A combination of recursive least squares and weighted least squares is made which can adapt its structure such that a relation between in- and output can he approximated, even when the structure of this relation is unknown beforehand. This method can adapt its structure on-line while it preserves information offered by previous samples, making it applicable in a control setting. This method has been tested with compntergenerated data, and it b used in a simulation to learn the non-linear state-dependent effects, both with good success.

Original language	English
Title of host publication	Proceedings of the 2003 IEEE International Symposium on Intelligent Control (ISIC)
Place of Publication	Piscataway, NJ
Publisher	OmniPress
Pages	75-80
Number of pages	5
ISBN (Print)	0-7803-7892-X, 0-7803-7891-1
DOIs	https://doi.org/10.1109/ISIC.2003.1253917
Publication status	Published - 5 Oct 2003
Event	2003 IEEE International Symposium on Intelligent Control, ISIC 2003 - Houston, United States Duration: 5 Oct 2003 → 8 Oct 2003

Publication series

Name	Proceedings of the IEEE International Symposium on Intelligent Control
Publisher	IEEE
Volume	2003
ISSN (Print)	2158-9860

Conference

Conference	2003 IEEE International Symposium on Intelligent Control, ISIC 2003
Abbreviated title	ISIC
Country/Territory	United States
City	Houston
Period	5/10/03 → 8/10/03

Access to Document

10.1109/ISIC.2003.1253917

Cite this

@inproceedings{51c47e286de345d09e404d445ed06963,

title = "On-line Nonparametric Regression to learn State-Dependent Disturbances",

abstract = "A combination of recursive least squares and weighted least squares is made which can adapt its structure such that a relation between in- and output can he approximated, even when the structure of this relation is unknown beforehand. This method can adapt its structure on-line while it preserves information offered by previous samples, making it applicable in a control setting. This method has been tested with compntergenerated data, and it b used in a simulation to learn the non-linear state-dependent effects, both with good success.",

author = "{de Kruif}, {Bas J.} and {de Vries}, {Theo J.A.}",

year = "2003",

month = oct,

day = "5",

doi = "10.1109/ISIC.2003.1253917",

language = "English",

isbn = "0-7803-7892-X",

series = "Proceedings of the IEEE International Symposium on Intelligent Control",

publisher = "OmniPress",

pages = "75--80",

booktitle = "Proceedings of the 2003 IEEE International Symposium on Intelligent Control (ISIC)",

note = "2003 IEEE International Symposium on Intelligent Control, ISIC 2003, ISIC ; Conference date: 05-10-2003 Through 08-10-2003",

}

de Kruif, BJ & de Vries, TJA 2003, On-line Nonparametric Regression to learn State-Dependent Disturbances. in Proceedings of the 2003 IEEE International Symposium on Intelligent Control (ISIC). Proceedings of the IEEE International Symposium on Intelligent Control, vol. 2003, OmniPress, Piscataway, NJ, pp. 75-80, 2003 IEEE International Symposium on Intelligent Control, ISIC 2003, Houston, Texas, United States, 5/10/03. https://doi.org/10.1109/ISIC.2003.1253917

On-line Nonparametric Regression to learn State-Dependent Disturbances. / de Kruif, Bas J.; de Vries, Theo J.A.

Proceedings of the 2003 IEEE International Symposium on Intelligent Control (ISIC). Piscataway, NJ : OmniPress, 2003. p. 75-80 (Proceedings of the IEEE International Symposium on Intelligent Control; Vol. 2003).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - On-line Nonparametric Regression to learn State-Dependent Disturbances

AU - de Kruif, Bas J.

AU - de Vries, Theo J.A.

PY - 2003/10/5

Y1 - 2003/10/5

N2 - A combination of recursive least squares and weighted least squares is made which can adapt its structure such that a relation between in- and output can he approximated, even when the structure of this relation is unknown beforehand. This method can adapt its structure on-line while it preserves information offered by previous samples, making it applicable in a control setting. This method has been tested with compntergenerated data, and it b used in a simulation to learn the non-linear state-dependent effects, both with good success.

AB - A combination of recursive least squares and weighted least squares is made which can adapt its structure such that a relation between in- and output can he approximated, even when the structure of this relation is unknown beforehand. This method can adapt its structure on-line while it preserves information offered by previous samples, making it applicable in a control setting. This method has been tested with compntergenerated data, and it b used in a simulation to learn the non-linear state-dependent effects, both with good success.

U2 - 10.1109/ISIC.2003.1253917

DO - 10.1109/ISIC.2003.1253917

M3 - Conference contribution

SN - 0-7803-7892-X

SN - 0-7803-7891-1

T3 - Proceedings of the IEEE International Symposium on Intelligent Control

SP - 75

EP - 80

BT - Proceedings of the 2003 IEEE International Symposium on Intelligent Control (ISIC)

PB - OmniPress

CY - Piscataway, NJ

T2 - 2003 IEEE International Symposium on Intelligent Control, ISIC 2003

Y2 - 5 October 2003 through 8 October 2003

ER -