Penalized ensemble Kalman filters for high dimensional non-linear systems.

The ensemble Kalman filter (EnKF) is a data assimilation technique that uses an ensemble of models, updated with data, to track the time evolution of a usually non-linear system. It does so by using an empirical approximation to the well-known Kalman filter. However, its performance can suffer when...

Full description

Saved in:
Bibliographic Details
Main Authors: Elizabeth Hou, Earl Lawrence, Alfred O Hero
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2021-01-01
Series:PLoS ONE
Online Access:https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0248046&type=printable
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850042712170954752
author Elizabeth Hou
Earl Lawrence
Alfred O Hero
author_facet Elizabeth Hou
Earl Lawrence
Alfred O Hero
author_sort Elizabeth Hou
collection DOAJ
description The ensemble Kalman filter (EnKF) is a data assimilation technique that uses an ensemble of models, updated with data, to track the time evolution of a usually non-linear system. It does so by using an empirical approximation to the well-known Kalman filter. However, its performance can suffer when the ensemble size is smaller than the state space, as is often necessary for computationally burdensome models. This scenario means that the empirical estimate of the state covariance is not full rank and possibly quite noisy. To solve this problem in this high dimensional regime, we propose a computationally fast and easy to implement algorithm called the penalized ensemble Kalman filter (PEnKF). Under certain conditions, it can be theoretically proven that the PEnKF will be accurate (the estimation error will converge to zero) despite having fewer ensemble members than state dimensions. Further, as contrasted to localization methods, the proposed approach learns the covariance structure associated with the dynamical system. These theoretical results are supported with simulations of several non-linear and high dimensional systems.
format Article
id doaj-art-4a3dae8e8db841369f9e7ecafd710aa6
institution DOAJ
issn 1932-6203
language English
publishDate 2021-01-01
publisher Public Library of Science (PLoS)
record_format Article
series PLoS ONE
spelling doaj-art-4a3dae8e8db841369f9e7ecafd710aa62025-08-20T02:55:28ZengPublic Library of Science (PLoS)PLoS ONE1932-62032021-01-01163e024804610.1371/journal.pone.0248046Penalized ensemble Kalman filters for high dimensional non-linear systems.Elizabeth HouEarl LawrenceAlfred O HeroThe ensemble Kalman filter (EnKF) is a data assimilation technique that uses an ensemble of models, updated with data, to track the time evolution of a usually non-linear system. It does so by using an empirical approximation to the well-known Kalman filter. However, its performance can suffer when the ensemble size is smaller than the state space, as is often necessary for computationally burdensome models. This scenario means that the empirical estimate of the state covariance is not full rank and possibly quite noisy. To solve this problem in this high dimensional regime, we propose a computationally fast and easy to implement algorithm called the penalized ensemble Kalman filter (PEnKF). Under certain conditions, it can be theoretically proven that the PEnKF will be accurate (the estimation error will converge to zero) despite having fewer ensemble members than state dimensions. Further, as contrasted to localization methods, the proposed approach learns the covariance structure associated with the dynamical system. These theoretical results are supported with simulations of several non-linear and high dimensional systems.https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0248046&type=printable
spellingShingle Elizabeth Hou
Earl Lawrence
Alfred O Hero
Penalized ensemble Kalman filters for high dimensional non-linear systems.
PLoS ONE
title Penalized ensemble Kalman filters for high dimensional non-linear systems.
title_full Penalized ensemble Kalman filters for high dimensional non-linear systems.
title_fullStr Penalized ensemble Kalman filters for high dimensional non-linear systems.
title_full_unstemmed Penalized ensemble Kalman filters for high dimensional non-linear systems.
title_short Penalized ensemble Kalman filters for high dimensional non-linear systems.
title_sort penalized ensemble kalman filters for high dimensional non linear systems
url https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0248046&type=printable
work_keys_str_mv AT elizabethhou penalizedensemblekalmanfiltersforhighdimensionalnonlinearsystems
AT earllawrence penalizedensemblekalmanfiltersforhighdimensionalnonlinearsystems
AT alfredohero penalizedensemblekalmanfiltersforhighdimensionalnonlinearsystems