Constrained kalman filtering

Improving fused information retention during constraining

Felix Baker, Suresh Thennadil

Research output: Chapter in Book/Report/Conference proceedingConference Paper published in ProceedingsResearchpeer-review

Abstract

Kalman filtering can produce unrealistic values and can prevent accurate convergence as the technique does not naturally include safeguards that exclude unphysical states. It can be demonstrated that without implementing constraints, or even some existing constraint strategies, that the filter could converge incorrectly. Currently available approaches to constraining the estimated state variables are arbitrary. For example, a simple way to constrain a violating state variable, is to reset its value to the constraint limit, the effect of which is a reduction of the importance of the measurement. The proposed constraining method attempts to preserve the importance of the observation/measurement in the fused estimate. This method compensates the changes in the constrained state variables by adjusting the non-constrained state variables in order to force the net change in measurement estimate to zero. The approach is implemented for the extended Kalman filters. The method is using a gas phase reaction in a Continuously Stirred Tank Reactor, with the state variables consisting of three species concentrations and the measurement is a pressure measurement with a known relationship to the state variables. The performance of the method is compared to currently available constraining techniques.

Original languageEnglish
Title of host publication2019 24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages434-437
Number of pages4
ISBN (Electronic)9781728109336
DOIs
Publication statusPublished - 1 Aug 2019
Event24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019 - Miedzyzdroje, Poland
Duration: 26 Aug 201929 Aug 2019

Publication series

Name2019 24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019

Conference

Conference24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019
CountryPoland
CityMiedzyzdroje
Period26/08/1929/08/19

Fingerprint

Kalman Filtering
Extended Kalman filters
Pressure measurement
Estimate
Kalman Filter
Reactor
Gases
Filter
Converge
Zero
Arbitrary

Cite this

Baker, F., & Thennadil, S. (2019). Constrained kalman filtering: Improving fused information retention during constraining. In 2019 24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019 (pp. 434-437). [8864655] (2019 24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019). IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/MMAR.2019.8864655
Baker, Felix ; Thennadil, Suresh. / Constrained kalman filtering : Improving fused information retention during constraining. 2019 24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019. IEEE, Institute of Electrical and Electronics Engineers, 2019. pp. 434-437 (2019 24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019).
@inproceedings{3f757aa125e64bc297d0418881fbfc48,
title = "Constrained kalman filtering: Improving fused information retention during constraining",
abstract = "Kalman filtering can produce unrealistic values and can prevent accurate convergence as the technique does not naturally include safeguards that exclude unphysical states. It can be demonstrated that without implementing constraints, or even some existing constraint strategies, that the filter could converge incorrectly. Currently available approaches to constraining the estimated state variables are arbitrary. For example, a simple way to constrain a violating state variable, is to reset its value to the constraint limit, the effect of which is a reduction of the importance of the measurement. The proposed constraining method attempts to preserve the importance of the observation/measurement in the fused estimate. This method compensates the changes in the constrained state variables by adjusting the non-constrained state variables in order to force the net change in measurement estimate to zero. The approach is implemented for the extended Kalman filters. The method is using a gas phase reaction in a Continuously Stirred Tank Reactor, with the state variables consisting of three species concentrations and the measurement is a pressure measurement with a known relationship to the state variables. The performance of the method is compared to currently available constraining techniques.",
keywords = "Compensated constraining, Constraint strategy, Extended Kalman filtering, Kalman filtering, Tracking",
author = "Felix Baker and Suresh Thennadil",
year = "2019",
month = "8",
day = "1",
doi = "10.1109/MMAR.2019.8864655",
language = "English",
series = "2019 24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
pages = "434--437",
booktitle = "2019 24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019",
address = "United States",

}

Baker, F & Thennadil, S 2019, Constrained kalman filtering: Improving fused information retention during constraining. in 2019 24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019., 8864655, 2019 24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019, IEEE, Institute of Electrical and Electronics Engineers, pp. 434-437, 24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019, Miedzyzdroje, Poland, 26/08/19. https://doi.org/10.1109/MMAR.2019.8864655

Constrained kalman filtering : Improving fused information retention during constraining. / Baker, Felix; Thennadil, Suresh.

2019 24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019. IEEE, Institute of Electrical and Electronics Engineers, 2019. p. 434-437 8864655 (2019 24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019).

Research output: Chapter in Book/Report/Conference proceedingConference Paper published in ProceedingsResearchpeer-review

TY - GEN

T1 - Constrained kalman filtering

T2 - Improving fused information retention during constraining

AU - Baker, Felix

AU - Thennadil, Suresh

PY - 2019/8/1

Y1 - 2019/8/1

N2 - Kalman filtering can produce unrealistic values and can prevent accurate convergence as the technique does not naturally include safeguards that exclude unphysical states. It can be demonstrated that without implementing constraints, or even some existing constraint strategies, that the filter could converge incorrectly. Currently available approaches to constraining the estimated state variables are arbitrary. For example, a simple way to constrain a violating state variable, is to reset its value to the constraint limit, the effect of which is a reduction of the importance of the measurement. The proposed constraining method attempts to preserve the importance of the observation/measurement in the fused estimate. This method compensates the changes in the constrained state variables by adjusting the non-constrained state variables in order to force the net change in measurement estimate to zero. The approach is implemented for the extended Kalman filters. The method is using a gas phase reaction in a Continuously Stirred Tank Reactor, with the state variables consisting of three species concentrations and the measurement is a pressure measurement with a known relationship to the state variables. The performance of the method is compared to currently available constraining techniques.

AB - Kalman filtering can produce unrealistic values and can prevent accurate convergence as the technique does not naturally include safeguards that exclude unphysical states. It can be demonstrated that without implementing constraints, or even some existing constraint strategies, that the filter could converge incorrectly. Currently available approaches to constraining the estimated state variables are arbitrary. For example, a simple way to constrain a violating state variable, is to reset its value to the constraint limit, the effect of which is a reduction of the importance of the measurement. The proposed constraining method attempts to preserve the importance of the observation/measurement in the fused estimate. This method compensates the changes in the constrained state variables by adjusting the non-constrained state variables in order to force the net change in measurement estimate to zero. The approach is implemented for the extended Kalman filters. The method is using a gas phase reaction in a Continuously Stirred Tank Reactor, with the state variables consisting of three species concentrations and the measurement is a pressure measurement with a known relationship to the state variables. The performance of the method is compared to currently available constraining techniques.

KW - Compensated constraining

KW - Constraint strategy

KW - Extended Kalman filtering

KW - Kalman filtering

KW - Tracking

UR - http://www.scopus.com/inward/record.url?scp=85074234710&partnerID=8YFLogxK

U2 - 10.1109/MMAR.2019.8864655

DO - 10.1109/MMAR.2019.8864655

M3 - Conference Paper published in Proceedings

T3 - 2019 24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019

SP - 434

EP - 437

BT - 2019 24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019

PB - IEEE, Institute of Electrical and Electronics Engineers

ER -

Baker F, Thennadil S. Constrained kalman filtering: Improving fused information retention during constraining. In 2019 24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019. IEEE, Institute of Electrical and Electronics Engineers. 2019. p. 434-437. 8864655. (2019 24th International Conference on Methods and Models in Automation and Robotics, MMAR 2019). https://doi.org/10.1109/MMAR.2019.8864655