Identification of switched autoregressive systems from large noisy data sets

Sarah Hojjatinia, Constantino Manuel Lagoa, Fabrizio Dabbene

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The paper introduces a novel methodology for the identification of coefficients of switched autoregressive linear models. We consider the case when the system's outputs are contaminated by possibly large values of measurement noise. It is assumed that only partial information on the probability distribution of the noise is available. Given input-output data, we aim at identifying switched system coefficients and parameters of the distribution of the noise which are compatible with the collected data. System dynamics are estimated through expected values computation and by exploiting the strong law of large numbers. We demonstrate the efficiency of the proposed approach with several academic examples. The method is shown to be extremely effective in the situations where a large number of measurements is available; cases in which previous approaches based on polynomial or mixed-integer optimization cannot be applied due to very large computational burden.

Original languageEnglish (US)
Title of host publication2019 American Control Conference, ACC 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages4313-4319
Number of pages7
ISBN (Electronic)9781538679265
StatePublished - Jul 1 2019
Event2019 American Control Conference, ACC 2019 - Philadelphia, United States
Duration: Jul 10 2019Jul 12 2019

Publication series

NameProceedings of the American Control Conference
Volume2019-July
ISSN (Print)0743-1619

Conference

Conference2019 American Control Conference, ACC 2019
CountryUnited States
CityPhiladelphia
Period7/10/197/12/19

Fingerprint

Probability distributions
Identification (control systems)
Dynamical systems
Polynomials

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering

Cite this

Hojjatinia, S., Lagoa, C. M., & Dabbene, F. (2019). Identification of switched autoregressive systems from large noisy data sets. In 2019 American Control Conference, ACC 2019 (pp. 4313-4319). [8814621] (Proceedings of the American Control Conference; Vol. 2019-July). Institute of Electrical and Electronics Engineers Inc..
Hojjatinia, Sarah ; Lagoa, Constantino Manuel ; Dabbene, Fabrizio. / Identification of switched autoregressive systems from large noisy data sets. 2019 American Control Conference, ACC 2019. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 4313-4319 (Proceedings of the American Control Conference).
@inproceedings{3677128c0ef14bbfb03efae92dc5b2d7,
title = "Identification of switched autoregressive systems from large noisy data sets",
abstract = "The paper introduces a novel methodology for the identification of coefficients of switched autoregressive linear models. We consider the case when the system's outputs are contaminated by possibly large values of measurement noise. It is assumed that only partial information on the probability distribution of the noise is available. Given input-output data, we aim at identifying switched system coefficients and parameters of the distribution of the noise which are compatible with the collected data. System dynamics are estimated through expected values computation and by exploiting the strong law of large numbers. We demonstrate the efficiency of the proposed approach with several academic examples. The method is shown to be extremely effective in the situations where a large number of measurements is available; cases in which previous approaches based on polynomial or mixed-integer optimization cannot be applied due to very large computational burden.",
author = "Sarah Hojjatinia and Lagoa, {Constantino Manuel} and Fabrizio Dabbene",
year = "2019",
month = "7",
day = "1",
language = "English (US)",
series = "Proceedings of the American Control Conference",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "4313--4319",
booktitle = "2019 American Control Conference, ACC 2019",
address = "United States",

}

Hojjatinia, S, Lagoa, CM & Dabbene, F 2019, Identification of switched autoregressive systems from large noisy data sets. in 2019 American Control Conference, ACC 2019., 8814621, Proceedings of the American Control Conference, vol. 2019-July, Institute of Electrical and Electronics Engineers Inc., pp. 4313-4319, 2019 American Control Conference, ACC 2019, Philadelphia, United States, 7/10/19.

Identification of switched autoregressive systems from large noisy data sets. / Hojjatinia, Sarah; Lagoa, Constantino Manuel; Dabbene, Fabrizio.

2019 American Control Conference, ACC 2019. Institute of Electrical and Electronics Engineers Inc., 2019. p. 4313-4319 8814621 (Proceedings of the American Control Conference; Vol. 2019-July).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Identification of switched autoregressive systems from large noisy data sets

AU - Hojjatinia, Sarah

AU - Lagoa, Constantino Manuel

AU - Dabbene, Fabrizio

PY - 2019/7/1

Y1 - 2019/7/1

N2 - The paper introduces a novel methodology for the identification of coefficients of switched autoregressive linear models. We consider the case when the system's outputs are contaminated by possibly large values of measurement noise. It is assumed that only partial information on the probability distribution of the noise is available. Given input-output data, we aim at identifying switched system coefficients and parameters of the distribution of the noise which are compatible with the collected data. System dynamics are estimated through expected values computation and by exploiting the strong law of large numbers. We demonstrate the efficiency of the proposed approach with several academic examples. The method is shown to be extremely effective in the situations where a large number of measurements is available; cases in which previous approaches based on polynomial or mixed-integer optimization cannot be applied due to very large computational burden.

AB - The paper introduces a novel methodology for the identification of coefficients of switched autoregressive linear models. We consider the case when the system's outputs are contaminated by possibly large values of measurement noise. It is assumed that only partial information on the probability distribution of the noise is available. Given input-output data, we aim at identifying switched system coefficients and parameters of the distribution of the noise which are compatible with the collected data. System dynamics are estimated through expected values computation and by exploiting the strong law of large numbers. We demonstrate the efficiency of the proposed approach with several academic examples. The method is shown to be extremely effective in the situations where a large number of measurements is available; cases in which previous approaches based on polynomial or mixed-integer optimization cannot be applied due to very large computational burden.

UR - http://www.scopus.com/inward/record.url?scp=85072274277&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072274277&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85072274277

T3 - Proceedings of the American Control Conference

SP - 4313

EP - 4319

BT - 2019 American Control Conference, ACC 2019

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Hojjatinia S, Lagoa CM, Dabbene F. Identification of switched autoregressive systems from large noisy data sets. In 2019 American Control Conference, ACC 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 4313-4319. 8814621. (Proceedings of the American Control Conference).