UNCERTAINTY IN DATA-DRIVEN KALMAN FILTERING FOR PARTIALLY KNOWN STATE-SPACE MODELS

Itzik Klein; Guy Revach; Nir Shlezinger; Jonas E. Mehr; Ruud J.G. van Sloun; Yonina C. Eldar

doi:10.1109/ICASSP43922.2022.9746732

UNCERTAINTY IN DATA-DRIVEN KALMAN FILTERING FOR PARTIALLY KNOWN STATE-SPACE MODELS

Itzik Klein, Guy Revach, Nir Shlezinger, Jonas E. Mehr, Ruud J.G. van Sloun, Yonina C. Eldar

Department of Marine Technologies

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

Abstract

Providing a metric of uncertainty alongside a state estimate is often crucial when tracking a dynamical system. Classic state estimators, such as the Kalman filter (KF), provide a time-dependent uncertainty measure from knowledge of the underlying statistics; however, deep learning based tracking systems struggle to reliably characterize uncertainty. In this paper, we investigate the ability of KalmanNet, a recently proposed; hybrid; model-based; deep state tracking algorithm, to estimate an uncertainty measure. By exploiting the interpretable nature of KalmanNet, we show that the error covariance matrix can be computed based on its internal features, as an uncertainty measure. We demonstrate that when the system dynamics are known, KalmanNet-which learns its mapping from data without access to the statistics-provides uncertainty similar to that provided by the KF; and while in the presence of evolution model-mismatch, KalmanNet provides a more accurate error estimation.

Original language	English
Title of host publication	2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	3194-3198
Number of pages	5
ISBN (Electronic)	9781665405409
DOIs	https://doi.org/10.1109/ICASSP43922.2022.9746732
State	Published - 2022
Event	47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Virtual, Online, Singapore Duration: 23 May 2022 → 27 May 2022

Publication series

Name	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume	2022-May
ISSN (Print)	1520-6149

Conference

Conference	47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022
Country/Territory	Singapore
City	Virtual, Online
Period	23/05/22 → 27/05/22

Bibliographical note

Publisher Copyright:
© 2022 IEEE

Keywords

Kalman filter
deep learning
uncertainty

ASJC Scopus subject areas

Software
Signal Processing
Electrical and Electronic Engineering

Access to Document

10.1109/ICASSP43922.2022.9746732

Cite this

Klein, I., Revach, G., Shlezinger, N., Mehr, J. E., van Sloun, R. J. G., & Eldar, Y. C. (2022). UNCERTAINTY IN DATA-DRIVEN KALMAN FILTERING FOR PARTIALLY KNOWN STATE-SPACE MODELS. In 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings (pp. 3194-3198). (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2022-May). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP43922.2022.9746732

Klein, Itzik ; Revach, Guy ; Shlezinger, Nir et al. / UNCERTAINTY IN DATA-DRIVEN KALMAN FILTERING FOR PARTIALLY KNOWN STATE-SPACE MODELS. 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2022. pp. 3194-3198 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

@inproceedings{e045748002d5421dab6d845a71ba3fc1,

title = "UNCERTAINTY IN DATA-DRIVEN KALMAN FILTERING FOR PARTIALLY KNOWN STATE-SPACE MODELS",

abstract = "Providing a metric of uncertainty alongside a state estimate is often crucial when tracking a dynamical system. Classic state estimators, such as the Kalman filter (KF), provide a time-dependent uncertainty measure from knowledge of the underlying statistics; however, deep learning based tracking systems struggle to reliably characterize uncertainty. In this paper, we investigate the ability of KalmanNet, a recently proposed; hybrid; model-based; deep state tracking algorithm, to estimate an uncertainty measure. By exploiting the interpretable nature of KalmanNet, we show that the error covariance matrix can be computed based on its internal features, as an uncertainty measure. We demonstrate that when the system dynamics are known, KalmanNet-which learns its mapping from data without access to the statistics-provides uncertainty similar to that provided by the KF; and while in the presence of evolution model-mismatch, KalmanNet provides a more accurate error estimation.",

keywords = "Kalman filter, deep learning, uncertainty",

author = "Itzik Klein and Guy Revach and Nir Shlezinger and Mehr, {Jonas E.} and {van Sloun}, {Ruud J.G.} and Eldar, {Yonina C.}",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE; 47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 ; Conference date: 23-05-2022 Through 27-05-2022",

year = "2022",

doi = "10.1109/ICASSP43922.2022.9746732",

language = "English",

series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "3194--3198",

booktitle = "2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings",

address = "United States",

}

Klein, I, Revach, G, Shlezinger, N, Mehr, JE, van Sloun, RJG & Eldar, YC 2022, UNCERTAINTY IN DATA-DRIVEN KALMAN FILTERING FOR PARTIALLY KNOWN STATE-SPACE MODELS. in 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, vol. 2022-May, Institute of Electrical and Electronics Engineers Inc., pp. 3194-3198, 47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022, Virtual, Online, Singapore, 23/05/22. https://doi.org/10.1109/ICASSP43922.2022.9746732

UNCERTAINTY IN DATA-DRIVEN KALMAN FILTERING FOR PARTIALLY KNOWN STATE-SPACE MODELS. / Klein, Itzik; Revach, Guy; Shlezinger, Nir et al.
2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2022. p. 3194-3198 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2022-May).

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

TY - GEN

T1 - UNCERTAINTY IN DATA-DRIVEN KALMAN FILTERING FOR PARTIALLY KNOWN STATE-SPACE MODELS

AU - Klein, Itzik

AU - Revach, Guy

AU - Shlezinger, Nir

AU - Mehr, Jonas E.

AU - van Sloun, Ruud J.G.

AU - Eldar, Yonina C.

PY - 2022

Y1 - 2022

N2 - Providing a metric of uncertainty alongside a state estimate is often crucial when tracking a dynamical system. Classic state estimators, such as the Kalman filter (KF), provide a time-dependent uncertainty measure from knowledge of the underlying statistics; however, deep learning based tracking systems struggle to reliably characterize uncertainty. In this paper, we investigate the ability of KalmanNet, a recently proposed; hybrid; model-based; deep state tracking algorithm, to estimate an uncertainty measure. By exploiting the interpretable nature of KalmanNet, we show that the error covariance matrix can be computed based on its internal features, as an uncertainty measure. We demonstrate that when the system dynamics are known, KalmanNet-which learns its mapping from data without access to the statistics-provides uncertainty similar to that provided by the KF; and while in the presence of evolution model-mismatch, KalmanNet provides a more accurate error estimation.

AB - Providing a metric of uncertainty alongside a state estimate is often crucial when tracking a dynamical system. Classic state estimators, such as the Kalman filter (KF), provide a time-dependent uncertainty measure from knowledge of the underlying statistics; however, deep learning based tracking systems struggle to reliably characterize uncertainty. In this paper, we investigate the ability of KalmanNet, a recently proposed; hybrid; model-based; deep state tracking algorithm, to estimate an uncertainty measure. By exploiting the interpretable nature of KalmanNet, we show that the error covariance matrix can be computed based on its internal features, as an uncertainty measure. We demonstrate that when the system dynamics are known, KalmanNet-which learns its mapping from data without access to the statistics-provides uncertainty similar to that provided by the KF; and while in the presence of evolution model-mismatch, KalmanNet provides a more accurate error estimation.

KW - Kalman filter

KW - deep learning

KW - uncertainty

UR - http://www.scopus.com/inward/record.url?scp=85128278542&partnerID=8YFLogxK

U2 - 10.1109/ICASSP43922.2022.9746732

DO - 10.1109/ICASSP43922.2022.9746732

M3 - Conference contribution

AN - SCOPUS:85128278542

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 3194

EP - 3198

BT - 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022

Y2 - 23 May 2022 through 27 May 2022

ER -

Klein I, Revach G, Shlezinger N, Mehr JE, van Sloun RJG, Eldar YC. UNCERTAINTY IN DATA-DRIVEN KALMAN FILTERING FOR PARTIALLY KNOWN STATE-SPACE MODELS. In 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2022. p. 3194-3198. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). doi: 10.1109/ICASSP43922.2022.9746732

UNCERTAINTY IN DATA-DRIVEN KALMAN FILTERING FOR PARTIALLY KNOWN STATE-SPACE MODELS

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this