Deep-Learning-Based Identification of LPV Models for Nonlinear Systems

Chris Verhoek; Gerben I. Beintema; Sofie Haesaert; Maarten Schoukens; Roland Tóth

doi:10.1109/CDC51059.2022.9992609

Deep-Learning-Based Identification of LPV Models for Nonlinear Systems

Control Systems

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

10 Citations (Scopus)

136 Downloads (Pure)

Abstract

The Linear Parameter-Varying (LPV) framework provides a modeling and control design toolchain to address nonlinear (NL) system behavior via linear surrogate models. Despite major research effort on LPV data-driven modeling, a key shortcoming of the current identification theory is that often the scheduling variable is assumed to be a given measured signal in the data set. In case of identifying an LPV model of a NL system, the selection of the scheduling map, which describes the relation to the measurable scheduling signal, is put on the users' shoulder, with only limited supporting tools available. This choice however greatly affects the usability and complexity of the resulting LPV model. This paper presents a deep-learning-based approach to provide joint estimation of a scheduling map and an LPV state-space model of a NL system from input-output data, and has consistency guarantees under general innovation-type noise conditions. Its efficiency is demonstrated on a realistic identification problem.

Original language	English
Title of host publication	61st IEEE Conference on Decision and Control
Publisher	Institute of Electrical and Electronics Engineers
Pages	3274-3280
Number of pages	7
ISBN (Electronic)	978-1-6654-6761-2
DOIs	https://doi.org/10.1109/CDC51059.2022.9992609
Publication status	Published - 10 Jan 2023
Event	61st IEEE Conference on Decision and Control, CDC 2022 - The Marriott Cancún Collection, Cancun, Mexico Duration: 6 Dec 2022 → 9 Dec 2022 Conference number: 61 https://cdc2022.ieeecss.org/

Conference

Conference	61st IEEE Conference on Decision and Control, CDC 2022
Abbreviated title	CDC 2022
Country/Territory	Mexico
City	Cancun
Period	6/12/22 → 9/12/22
Internet address	https://cdc2022.ieeecss.org/

Funding

This work was supported by the European Space Agency in the scope of the \u2018AI4GNC\u2019 project with SENER Aeroespacial S.A. (contract nr. 4000133595/20/NL/CRS) and was also supported by the European Union within the framework of the National Laboratory for Autonomous Systems (RRF-2.3.1-21-2022-00002).

Funders	Funder number
European Union's Horizon 2020 - Research and Innovation Framework Programme	714663

Keywords

Deep Learning
Linear Parameter-Varying Systems
Nonlinear Systems
System Identification

Access to Document

10.1109/CDC51059.2022.9992609

Final_Author_versionFinal author version , 1.43 MBLicence: CC BY-SA

Cite this

@inproceedings{7f9cd67b4a254953aae8a2e64e149988,

title = "Deep-Learning-Based Identification of LPV Models for Nonlinear Systems",

abstract = "The Linear Parameter-Varying (LPV) framework provides a modeling and control design toolchain to address nonlinear (NL) system behavior via linear surrogate models. Despite major research effort on LPV data-driven modeling, a key shortcoming of the current identification theory is that often the scheduling variable is assumed to be a given measured signal in the data set. In case of identifying an LPV model of a NL system, the selection of the scheduling map, which describes the relation to the measurable scheduling signal, is put on the users' shoulder, with only limited supporting tools available. This choice however greatly affects the usability and complexity of the resulting LPV model. This paper presents a deep-learning-based approach to provide joint estimation of a scheduling map and an LPV state-space model of a NL system from input-output data, and has consistency guarantees under general innovation-type noise conditions. Its efficiency is demonstrated on a realistic identification problem.",

keywords = "eess.SY, cs.SY, math.DS, math.OC, Deep Learning, Linear Parameter-Varying Systems, Nonlinear Systems, System Identification",

author = "Chris Verhoek and Beintema, \{Gerben I.\} and Sofie Haesaert and Maarten Schoukens and Roland T{\'o}th",

year = "2023",

month = jan,

day = "10",

doi = "10.1109/CDC51059.2022.9992609",

language = "English",

pages = "3274--3280",

booktitle = "61st IEEE Conference on Decision and Control",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "61st IEEE Conference on Decision and Control, CDC 2022, CDC 2022 ; Conference date: 06-12-2022 Through 09-12-2022",

url = "https://cdc2022.ieeecss.org/",

}

Verhoek, C , Beintema, GI , Haesaert, S , Schoukens, M & Tóth, R 2023, Deep-Learning-Based Identification of LPV Models for Nonlinear Systems. in 61st IEEE Conference on Decision and Control., 9992609, Institute of Electrical and Electronics Engineers, pp. 3274-3280, 61st IEEE Conference on Decision and Control, CDC 2022, Cancun, Mexico, 6/12/22. https://doi.org/10.1109/CDC51059.2022.9992609

Deep-Learning-Based Identification of LPV Models for Nonlinear Systems. / Verhoek, Chris ; Beintema, Gerben I.; Haesaert, Sofie et al.
61st IEEE Conference on Decision and Control. Institute of Electrical and Electronics Engineers, 2023. p. 3274-3280 9992609.

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

TY - GEN

T1 - Deep-Learning-Based Identification of LPV Models for Nonlinear Systems

AU - Verhoek, Chris

AU - Beintema, Gerben I.

AU - Haesaert, Sofie

AU - Schoukens, Maarten

AU - Tóth, Roland

N1 - Conference code: 61

PY - 2023/1/10

Y1 - 2023/1/10

N2 - The Linear Parameter-Varying (LPV) framework provides a modeling and control design toolchain to address nonlinear (NL) system behavior via linear surrogate models. Despite major research effort on LPV data-driven modeling, a key shortcoming of the current identification theory is that often the scheduling variable is assumed to be a given measured signal in the data set. In case of identifying an LPV model of a NL system, the selection of the scheduling map, which describes the relation to the measurable scheduling signal, is put on the users' shoulder, with only limited supporting tools available. This choice however greatly affects the usability and complexity of the resulting LPV model. This paper presents a deep-learning-based approach to provide joint estimation of a scheduling map and an LPV state-space model of a NL system from input-output data, and has consistency guarantees under general innovation-type noise conditions. Its efficiency is demonstrated on a realistic identification problem.

AB - The Linear Parameter-Varying (LPV) framework provides a modeling and control design toolchain to address nonlinear (NL) system behavior via linear surrogate models. Despite major research effort on LPV data-driven modeling, a key shortcoming of the current identification theory is that often the scheduling variable is assumed to be a given measured signal in the data set. In case of identifying an LPV model of a NL system, the selection of the scheduling map, which describes the relation to the measurable scheduling signal, is put on the users' shoulder, with only limited supporting tools available. This choice however greatly affects the usability and complexity of the resulting LPV model. This paper presents a deep-learning-based approach to provide joint estimation of a scheduling map and an LPV state-space model of a NL system from input-output data, and has consistency guarantees under general innovation-type noise conditions. Its efficiency is demonstrated on a realistic identification problem.

KW - eess.SY

KW - cs.SY

KW - math.DS

KW - math.OC

KW - Deep Learning

KW - Linear Parameter-Varying Systems

KW - Nonlinear Systems

KW - System Identification

UR - https://www.scopus.com/pages/publications/85147013741

U2 - 10.1109/CDC51059.2022.9992609

DO - 10.1109/CDC51059.2022.9992609

M3 - Conference contribution

SP - 3274

EP - 3280

BT - 61st IEEE Conference on Decision and Control

PB - Institute of Electrical and Electronics Engineers

T2 - 61st IEEE Conference on Decision and Control, CDC 2022

Y2 - 6 December 2022 through 9 December 2022

ER -

Deep-Learning-Based Identification of LPV Models for Nonlinear Systems

Abstract

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this