Comparison of covariance estimation using autocovariance LS method and adaptive SRUKF

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Authors

  • Mauro Hernan Riva
  • Matthias Dagen
  • Tobias Ortmaier

Research Organisations

View graph of relations

Details

Original languageEnglish
Title of host publication2017 American Control Conference, ACC 2017
Place of PublicationSeattle, USA
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages5780-5786
Number of pages7
ISBN (electronic)9781509059928
Publication statusPublished - 29 Jun 2017
Event2017 American Control Conference, ACC 2017 - Seattle, United States
Duration: 24 May 201726 May 2017

Publication series

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

Abstract

State estimators are used to reconstruct current plant states based on information received from plant sensors and the use of a mathematical model. The typically applied Kalman filter derivatives require knowledge about the noise statistics affecting system states and measurements. These are often unknown and inaccurate parameterization may lead to decreased filter performance or even filter divergence. In this paper, a comparison between two covariance estimation methods is presented. The offline time-varying autocovariance Least-Square (LS) method is compared to the online adaptive Square-Root Unscented Kalman Filter (SRUKF). Both methods are evaluated in simulations and experiments using a pendubot w.r.t. robustness against random covariance initializations and state estimation accuracy. The results show that with both methods the filter performance can remarkably be improved.

ASJC Scopus subject areas

Cite this

Comparison of covariance estimation using autocovariance LS method and adaptive SRUKF. / Riva, Mauro Hernan; Dagen, Matthias; Ortmaier, Tobias.
2017 American Control Conference, ACC 2017. Seattle, USA: Institute of Electrical and Electronics Engineers Inc., 2017. p. 5780-5786 7963856 (Proceedings of the American Control Conference).

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Riva, MH, Dagen, M & Ortmaier, T 2017, Comparison of covariance estimation using autocovariance LS method and adaptive SRUKF. in 2017 American Control Conference, ACC 2017., 7963856, Proceedings of the American Control Conference, Institute of Electrical and Electronics Engineers Inc., Seattle, USA, pp. 5780-5786, 2017 American Control Conference, ACC 2017, Seattle, United States, 24 May 2017. https://doi.org/10.23919/acc.2017.7963856
Riva, M. H., Dagen, M., & Ortmaier, T. (2017). Comparison of covariance estimation using autocovariance LS method and adaptive SRUKF. In 2017 American Control Conference, ACC 2017 (pp. 5780-5786). Article 7963856 (Proceedings of the American Control Conference). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.23919/acc.2017.7963856
Riva MH, Dagen M, Ortmaier T. Comparison of covariance estimation using autocovariance LS method and adaptive SRUKF. In 2017 American Control Conference, ACC 2017. Seattle, USA: Institute of Electrical and Electronics Engineers Inc. 2017. p. 5780-5786. 7963856. (Proceedings of the American Control Conference). doi: 10.23919/acc.2017.7963856
Riva, Mauro Hernan ; Dagen, Matthias ; Ortmaier, Tobias. / Comparison of covariance estimation using autocovariance LS method and adaptive SRUKF. 2017 American Control Conference, ACC 2017. Seattle, USA : Institute of Electrical and Electronics Engineers Inc., 2017. pp. 5780-5786 (Proceedings of the American Control Conference).
Download
@inproceedings{d4268bf01aec4c60a68a3c2ea38bfdcf,
title = "Comparison of covariance estimation using autocovariance LS method and adaptive SRUKF",
abstract = "State estimators are used to reconstruct current plant states based on information received from plant sensors and the use of a mathematical model. The typically applied Kalman filter derivatives require knowledge about the noise statistics affecting system states and measurements. These are often unknown and inaccurate parameterization may lead to decreased filter performance or even filter divergence. In this paper, a comparison between two covariance estimation methods is presented. The offline time-varying autocovariance Least-Square (LS) method is compared to the online adaptive Square-Root Unscented Kalman Filter (SRUKF). Both methods are evaluated in simulations and experiments using a pendubot w.r.t. robustness against random covariance initializations and state estimation accuracy. The results show that with both methods the filter performance can remarkably be improved.",
author = "Riva, {Mauro Hernan} and Matthias Dagen and Tobias Ortmaier",
year = "2017",
month = jun,
day = "29",
doi = "10.23919/acc.2017.7963856",
language = "English",
series = "Proceedings of the American Control Conference",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "5780--5786",
booktitle = "2017 American Control Conference, ACC 2017",
address = "United States",
note = "2017 American Control Conference, ACC 2017 ; Conference date: 24-05-2017 Through 26-05-2017",

}

Download

TY - GEN

T1 - Comparison of covariance estimation using autocovariance LS method and adaptive SRUKF

AU - Riva, Mauro Hernan

AU - Dagen, Matthias

AU - Ortmaier, Tobias

PY - 2017/6/29

Y1 - 2017/6/29

N2 - State estimators are used to reconstruct current plant states based on information received from plant sensors and the use of a mathematical model. The typically applied Kalman filter derivatives require knowledge about the noise statistics affecting system states and measurements. These are often unknown and inaccurate parameterization may lead to decreased filter performance or even filter divergence. In this paper, a comparison between two covariance estimation methods is presented. The offline time-varying autocovariance Least-Square (LS) method is compared to the online adaptive Square-Root Unscented Kalman Filter (SRUKF). Both methods are evaluated in simulations and experiments using a pendubot w.r.t. robustness against random covariance initializations and state estimation accuracy. The results show that with both methods the filter performance can remarkably be improved.

AB - State estimators are used to reconstruct current plant states based on information received from plant sensors and the use of a mathematical model. The typically applied Kalman filter derivatives require knowledge about the noise statistics affecting system states and measurements. These are often unknown and inaccurate parameterization may lead to decreased filter performance or even filter divergence. In this paper, a comparison between two covariance estimation methods is presented. The offline time-varying autocovariance Least-Square (LS) method is compared to the online adaptive Square-Root Unscented Kalman Filter (SRUKF). Both methods are evaluated in simulations and experiments using a pendubot w.r.t. robustness against random covariance initializations and state estimation accuracy. The results show that with both methods the filter performance can remarkably be improved.

UR - http://www.scopus.com/inward/record.url?scp=85027005585&partnerID=8YFLogxK

U2 - 10.23919/acc.2017.7963856

DO - 10.23919/acc.2017.7963856

M3 - Conference contribution

AN - SCOPUS:85027005585

T3 - Proceedings of the American Control Conference

SP - 5780

EP - 5786

BT - 2017 American Control Conference, ACC 2017

PB - Institute of Electrical and Electronics Engineers Inc.

CY - Seattle, USA

T2 - 2017 American Control Conference, ACC 2017

Y2 - 24 May 2017 through 26 May 2017

ER -