Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

MSE-optimal measurement dimension reduction in gaussian filtering

Greiff, Marcus LU ; Robertsson, Anders LU and Berntorp, Karl LU (2020) 4th IEEE Conference on Control Technology and Applications, CCTA 2020 In CCTA 2020 - 4th IEEE Conference on Control Technology and Applications p.126-133
Abstract

We present a framework for measurement dimension reduction in Gaussian filtering, defined in terms of a linear operator acting on the measurement vector. This operator is optimized to minimize the Cramér-Rao bound of the estimate's mean squared error (MSE), yielding a measurement subspace from which elements minimally worsen the filter MSE performance, as compared to filtering with the original measurements. This is demonstrated with Kalman filtering in a linear Gaussian setting and various non-linear Gaussian filters with an on-line adaption of the operator. The proposed method improves computational time in exchange for a quantifiable and sometimes negligibly worsened MSE of the estimate.

Please use this url to cite or link to this publication:
author
; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
host publication
CCTA 2020 - 4th IEEE Conference on Control Technology and Applications
series title
CCTA 2020 - 4th IEEE Conference on Control Technology and Applications
article number
9206162
pages
8 pages
publisher
IEEE - Institute of Electrical and Electronics Engineers Inc.
conference name
4th IEEE Conference on Control Technology and Applications, CCTA 2020
conference location
Virtual, Montreal, Canada
conference dates
2020-08-24 - 2020-08-26
external identifiers
  • scopus:85094159980
ISBN
9781728171401
DOI
10.1109/CCTA41146.2020.9206162
language
English
LU publication?
yes
id
82086b57-b7ee-431f-8f6f-5926e546aa70
date added to LUP
2020-11-10 12:12:40
date last changed
2022-05-12 07:41:05
@inproceedings{82086b57-b7ee-431f-8f6f-5926e546aa70,
  abstract     = {{<p>We present a framework for measurement dimension reduction in Gaussian filtering, defined in terms of a linear operator acting on the measurement vector. This operator is optimized to minimize the Cramér-Rao bound of the estimate's mean squared error (MSE), yielding a measurement subspace from which elements minimally worsen the filter MSE performance, as compared to filtering with the original measurements. This is demonstrated with Kalman filtering in a linear Gaussian setting and various non-linear Gaussian filters with an on-line adaption of the operator. The proposed method improves computational time in exchange for a quantifiable and sometimes negligibly worsened MSE of the estimate.</p>}},
  author       = {{Greiff, Marcus and Robertsson, Anders and Berntorp, Karl}},
  booktitle    = {{CCTA 2020 - 4th IEEE Conference on Control Technology and Applications}},
  isbn         = {{9781728171401}},
  language     = {{eng}},
  pages        = {{126--133}},
  publisher    = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}},
  series       = {{CCTA 2020 - 4th IEEE Conference on Control Technology and Applications}},
  title        = {{MSE-optimal measurement dimension reduction in gaussian filtering}},
  url          = {{http://dx.doi.org/10.1109/CCTA41146.2020.9206162}},
  doi          = {{10.1109/CCTA41146.2020.9206162}},
  year         = {{2020}},
}