Mobile sensor networks for learning anisotropic gaussian processes

Yunfei Xu, Jongeun Choi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

12 Citations (Scopus)

Abstract

This paper presents a novel class of self-organizing sensing agents that learn an anisotropic, spatio-temporal Gaussian process using noisy measurements and move in order to improve the quality of the estimated covariance function. This approach is based on a class of anisotropic covariance functions of Gaussian processes developed to model a broad range of anisotropic, spatio-temporal physical phenomena. The covariance function is assumed to be unknown a priori. Hence, it is estimated by the maximum likelihood (ML) estimator. The prediction of the field of interest is then obtained based on a non-parametric approach. An optimal navigation strategy is proposed to minimize the Cramér-Rao lower bound (CRLB) of the estimation error covariance matrix. Simulation results demonstrate the effectiveness of the proposed scheme.

Original languageEnglish
Title of host publication2009 American Control Conference, ACC 2009
Pages5049-5054
Number of pages6
DOIs
Publication statusPublished - 2009 Nov 23
Event2009 American Control Conference, ACC 2009 - St. Louis, MO, United States
Duration: 2009 Jun 102009 Jun 12

Publication series

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

Other

Other2009 American Control Conference, ACC 2009
CountryUnited States
CitySt. Louis, MO
Period09/6/1009/6/12

Fingerprint

Sensor networks
Wireless networks
Covariance matrix
Error analysis
Maximum likelihood
Navigation

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering

Cite this

Xu, Y., & Choi, J. (2009). Mobile sensor networks for learning anisotropic gaussian processes. In 2009 American Control Conference, ACC 2009 (pp. 5049-5054). [5160470] (Proceedings of the American Control Conference). https://doi.org/10.1109/ACC.2009.5160470
Xu, Yunfei ; Choi, Jongeun. / Mobile sensor networks for learning anisotropic gaussian processes. 2009 American Control Conference, ACC 2009. 2009. pp. 5049-5054 (Proceedings of the American Control Conference).
@inproceedings{22ffa49d6a10498eacd4ece49a4ee2fc,
title = "Mobile sensor networks for learning anisotropic gaussian processes",
abstract = "This paper presents a novel class of self-organizing sensing agents that learn an anisotropic, spatio-temporal Gaussian process using noisy measurements and move in order to improve the quality of the estimated covariance function. This approach is based on a class of anisotropic covariance functions of Gaussian processes developed to model a broad range of anisotropic, spatio-temporal physical phenomena. The covariance function is assumed to be unknown a priori. Hence, it is estimated by the maximum likelihood (ML) estimator. The prediction of the field of interest is then obtained based on a non-parametric approach. An optimal navigation strategy is proposed to minimize the Cram{\'e}r-Rao lower bound (CRLB) of the estimation error covariance matrix. Simulation results demonstrate the effectiveness of the proposed scheme.",
author = "Yunfei Xu and Jongeun Choi",
year = "2009",
month = "11",
day = "23",
doi = "10.1109/ACC.2009.5160470",
language = "English",
isbn = "9781424445240",
series = "Proceedings of the American Control Conference",
pages = "5049--5054",
booktitle = "2009 American Control Conference, ACC 2009",

}

Xu, Y & Choi, J 2009, Mobile sensor networks for learning anisotropic gaussian processes. in 2009 American Control Conference, ACC 2009., 5160470, Proceedings of the American Control Conference, pp. 5049-5054, 2009 American Control Conference, ACC 2009, St. Louis, MO, United States, 09/6/10. https://doi.org/10.1109/ACC.2009.5160470

Mobile sensor networks for learning anisotropic gaussian processes. / Xu, Yunfei; Choi, Jongeun.

2009 American Control Conference, ACC 2009. 2009. p. 5049-5054 5160470 (Proceedings of the American Control Conference).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Mobile sensor networks for learning anisotropic gaussian processes

AU - Xu, Yunfei

AU - Choi, Jongeun

PY - 2009/11/23

Y1 - 2009/11/23

N2 - This paper presents a novel class of self-organizing sensing agents that learn an anisotropic, spatio-temporal Gaussian process using noisy measurements and move in order to improve the quality of the estimated covariance function. This approach is based on a class of anisotropic covariance functions of Gaussian processes developed to model a broad range of anisotropic, spatio-temporal physical phenomena. The covariance function is assumed to be unknown a priori. Hence, it is estimated by the maximum likelihood (ML) estimator. The prediction of the field of interest is then obtained based on a non-parametric approach. An optimal navigation strategy is proposed to minimize the Cramér-Rao lower bound (CRLB) of the estimation error covariance matrix. Simulation results demonstrate the effectiveness of the proposed scheme.

AB - This paper presents a novel class of self-organizing sensing agents that learn an anisotropic, spatio-temporal Gaussian process using noisy measurements and move in order to improve the quality of the estimated covariance function. This approach is based on a class of anisotropic covariance functions of Gaussian processes developed to model a broad range of anisotropic, spatio-temporal physical phenomena. The covariance function is assumed to be unknown a priori. Hence, it is estimated by the maximum likelihood (ML) estimator. The prediction of the field of interest is then obtained based on a non-parametric approach. An optimal navigation strategy is proposed to minimize the Cramér-Rao lower bound (CRLB) of the estimation error covariance matrix. Simulation results demonstrate the effectiveness of the proposed scheme.

UR - http://www.scopus.com/inward/record.url?scp=70449643307&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70449643307&partnerID=8YFLogxK

U2 - 10.1109/ACC.2009.5160470

DO - 10.1109/ACC.2009.5160470

M3 - Conference contribution

AN - SCOPUS:70449643307

SN - 9781424445240

T3 - Proceedings of the American Control Conference

SP - 5049

EP - 5054

BT - 2009 American Control Conference, ACC 2009

ER -

Xu Y, Choi J. Mobile sensor networks for learning anisotropic gaussian processes. In 2009 American Control Conference, ACC 2009. 2009. p. 5049-5054. 5160470. (Proceedings of the American Control Conference). https://doi.org/10.1109/ACC.2009.5160470