Nonparametric Identification of Linear Time-Varying Systems using Gaussian Process Regression

Noel Hallemans; John Lataire; Rik Pintelon

doi:10.1016/j.ifacol.2020.12.1277

Nonparametric Identification of Linear Time-Varying Systems using Gaussian Process Regression

Noel Hallemans, John Lataire, Rik Pintelon

Research output: Chapter in Book/Report/Conference proceeding › Conference paper

4 Citations (Scopus)

92 Downloads (Pure)

Abstract

Linear time-varying systems are a class of systems, the dynamics of which evolve in time. This results in a time-varying frequency response function where each frequency has a time-varying gain. In classical identification techniques, basis functions are employed to fit these time-varying gains. In this paper a new method based on Gaussian process regression is presented. The advantage of the proposed method is a more convenient model structure and model order selection.

Original language	English
Title of host publication	IFAC-PapersOnLine
Editors	Rolf Findeisen, Sandra Hirche, Klaus Janschek, Martin Mönnigmann
Publisher	Elsevier
Pages	1001-1006
Number of pages	6
Volume	53
Edition	2
ISBN (Print)	2405-8963
DOIs	https://doi.org/10.1016/j.ifacol.2020.12.1277
Publication status	Published - 1 Jan 2020
Event	21st IFAC World Congress : Automatic Control – Meeting Societal Challenges - virtual IFAC World Congress, Berlin, Germany Duration: 11 Jul 2020 → 17 Jul 2020 https://www.ifac2020.org/welcome/

Publication series

Name	IFAC-PapersOnLine
ISSN (Print)	2405-8963

Conference

Conference	21st IFAC World Congress
Country/Territory	Germany
City	Berlin
Period	11/07/20 → 17/07/20
Internet address	https://www.ifac2020.org/welcome/

Bibliographical note

Funding Information:
This research was supported in part by the Fund for Scientific Research (FWO Vlaanderen), and in part by the Flemish Government (Methusalem Grant METH1).

Publisher Copyright:
Copyright © 2020 The Authors. This is an open access article under the CC BY-NC-ND license

Copyright:
Copyright 2022 Elsevier B.V., All rights reserved.

Keywords

Identification
Linear time-varying systems
Gaussian processes
Machine learning

Access to Document

10.1016/j.ifacol.2020.12.1277

LTVGP_IFACpaperNoelAccepted author manuscript, 1.02 MB

https://www.sciencedirect.com/science/article/pii/S2405896320316797

Cite this

@inproceedings{f5ce39343bfc461cad453efa794bea6d,

title = "Nonparametric Identification of Linear Time-Varying Systems using Gaussian Process Regression",

abstract = "Linear time-varying systems are a class of systems, the dynamics of which evolve in time. This results in a time-varying frequency response function where each frequency has a time-varying gain. In classical identification techniques, basis functions are employed to fit these time-varying gains. In this paper a new method based on Gaussian process regression is presented. The advantage of the proposed method is a more convenient model structure and model order selection.",

keywords = "Identification, Linear time-varying systems, Gaussian processes, Machine learning",

author = "Noel Hallemans and John Lataire and Rik Pintelon",

note = "Funding Information: This research was supported in part by the Fund for Scientific Research (FWO Vlaanderen), and in part by the Flemish Government (Methusalem Grant METH1). Publisher Copyright: Copyright {\textcopyright} 2020 The Authors. This is an open access article under the CC BY-NC-ND license Copyright: Copyright 2022 Elsevier B.V., All rights reserved.; 21st IFAC World Congress : Automatic Control – Meeting Societal Challenges ; Conference date: 11-07-2020 Through 17-07-2020",

year = "2020",

month = jan,

day = "1",

doi = "10.1016/j.ifacol.2020.12.1277",

language = "English",

isbn = "2405-8963",

volume = "53",

series = "IFAC-PapersOnLine",

publisher = "Elsevier",

pages = "1001--1006",

editor = "Rolf Findeisen and Sandra Hirche and Klaus Janschek and Martin M{\"o}nnigmann",

booktitle = "IFAC-PapersOnLine",

edition = "2",

url = "https://www.ifac2020.org/welcome/",

}

Hallemans, N, Lataire, J & Pintelon, R 2020, Nonparametric Identification of Linear Time-Varying Systems using Gaussian Process Regression. in R Findeisen, S Hirche, K Janschek & M Mönnigmann (eds), IFAC-PapersOnLine. 2 edn, vol. 53, IFAC-PapersOnLine, Elsevier, pp. 1001-1006, 21st IFAC World Congress , Berlin, Germany, 11/07/20. https://doi.org/10.1016/j.ifacol.2020.12.1277

Nonparametric Identification of Linear Time-Varying Systems using Gaussian Process Regression. / Hallemans, Noel; Lataire, John ; Pintelon, Rik.

IFAC-PapersOnLine. ed. / Rolf Findeisen; Sandra Hirche; Klaus Janschek; Martin Mönnigmann. Vol. 53 2. ed. Elsevier, 2020. p. 1001-1006 (IFAC-PapersOnLine).

Research output: Chapter in Book/Report/Conference proceeding › Conference paper

TY - GEN

T1 - Nonparametric Identification of Linear Time-Varying Systems using Gaussian Process Regression

AU - Hallemans, Noel

AU - Lataire, John

AU - Pintelon, Rik

N1 - Funding Information: This research was supported in part by the Fund for Scientific Research (FWO Vlaanderen), and in part by the Flemish Government (Methusalem Grant METH1). Publisher Copyright: Copyright © 2020 The Authors. This is an open access article under the CC BY-NC-ND license Copyright: Copyright 2022 Elsevier B.V., All rights reserved.

PY - 2020/1/1

Y1 - 2020/1/1

N2 - Linear time-varying systems are a class of systems, the dynamics of which evolve in time. This results in a time-varying frequency response function where each frequency has a time-varying gain. In classical identification techniques, basis functions are employed to fit these time-varying gains. In this paper a new method based on Gaussian process regression is presented. The advantage of the proposed method is a more convenient model structure and model order selection.

AB - Linear time-varying systems are a class of systems, the dynamics of which evolve in time. This results in a time-varying frequency response function where each frequency has a time-varying gain. In classical identification techniques, basis functions are employed to fit these time-varying gains. In this paper a new method based on Gaussian process regression is presented. The advantage of the proposed method is a more convenient model structure and model order selection.

KW - Identification

KW - Linear time-varying systems

KW - Gaussian processes

KW - Machine learning

UR - http://www.scopus.com/inward/record.url?scp=85105058296&partnerID=8YFLogxK

U2 - 10.1016/j.ifacol.2020.12.1277

DO - 10.1016/j.ifacol.2020.12.1277

M3 - Conference paper

SN - 2405-8963

VL - 53

T3 - IFAC-PapersOnLine

SP - 1001

EP - 1006

BT - IFAC-PapersOnLine

A2 - Findeisen, Rolf

A2 - Hirche, Sandra

A2 - Janschek, Klaus

A2 - Mönnigmann, Martin

PB - Elsevier

T2 - 21st IFAC World Congress

Y2 - 11 July 2020 through 17 July 2020

ER -

Nonparametric Identification of Linear Time-Varying Systems using Gaussian Process Regression

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this