Model Fitting Filter

Xiaoxu Wang; Jun Zhang; Chaofeng Li; Quan Pan; Zhengtao Ding; Xin Yin

doi:10.23919/ACC.2018.8431644

Model Fitting Filter

Xiaoxu Wang, Jun Zhang, Chaofeng Li, Quan Pan, Zhengtao Ding, Xin Yin

School of Automation

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

Abstract

In this paper, we present a new nonlinear filter based on variational inference, which we have named the model fitting filter (MFF). The core of MFF is to represent or fit the nonlinear measurement likelihood probability with a linear Gaussian probability (LGP) characterized by some parameters that need to be turned. Further, MFF is analytically established on LGP, which is a little similar to the simple linear system filter. Specifically, the nonlinear filtering performance is well improved by utilizing these turned parameters to correct the state estimation. At the same time, MFF does not require the square root decomposition of covariance matrix, so it avoids the inborn defect of numerical instability, which exists more or less in UKF or DDF. The superior performance of MFF is numerically tested by a target tracking problem with nonlinear models.

Original language	English
Title of host publication	2018 Annual American Control Conference, ACC 2018
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	5479-5484
Number of pages	6
ISBN (Print)	9781538654286
DOIs	https://doi.org/10.23919/ACC.2018.8431644
State	Published - 9 Aug 2018
Event	2018 Annual American Control Conference, ACC 2018 - Milwauke, United States Duration: 27 Jun 2018 → 29 Jun 2018

Publication series

Name	Proceedings of the American Control Conference
Volume	2018-June
ISSN (Print)	0743-1619

Conference

Conference	2018 Annual American Control Conference, ACC 2018
Country/Territory	United States
City	Milwauke
Period	27/06/18 → 29/06/18

Keywords

model fitting filter
Nonlinear estimation
numerical instability
variational inference

Access to Document

10.23919/ACC.2018.8431644

Cite this

@inproceedings{482a9109bdbf4786ae46a9b6c95f1052,

title = "Model Fitting Filter",

abstract = "In this paper, we present a new nonlinear filter based on variational inference, which we have named the model fitting filter (MFF). The core of MFF is to represent or fit the nonlinear measurement likelihood probability with a linear Gaussian probability (LGP) characterized by some parameters that need to be turned. Further, MFF is analytically established on LGP, which is a little similar to the simple linear system filter. Specifically, the nonlinear filtering performance is well improved by utilizing these turned parameters to correct the state estimation. At the same time, MFF does not require the square root decomposition of covariance matrix, so it avoids the inborn defect of numerical instability, which exists more or less in UKF or DDF. The superior performance of MFF is numerically tested by a target tracking problem with nonlinear models.",

keywords = "model fitting filter, Nonlinear estimation, numerical instability, variational inference",

author = "Xiaoxu Wang and Jun Zhang and Chaofeng Li and Quan Pan and Zhengtao Ding and Xin Yin",

note = "Publisher Copyright: {\textcopyright} 2018 AACC.; 2018 Annual American Control Conference, ACC 2018 ; Conference date: 27-06-2018 Through 29-06-2018",

year = "2018",

month = aug,

day = "9",

doi = "10.23919/ACC.2018.8431644",

language = "英语",

isbn = "9781538654286",

series = "Proceedings of the American Control Conference",

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

pages = "5479--5484",

booktitle = "2018 Annual American Control Conference, ACC 2018",

}

Wang, X, Zhang, J, Li, C, Pan, Q, Ding, Z & Yin, X 2018, Model Fitting Filter. in 2018 Annual American Control Conference, ACC 2018., 8431644, Proceedings of the American Control Conference, vol. 2018-June, Institute of Electrical and Electronics Engineers Inc., pp. 5479-5484, 2018 Annual American Control Conference, ACC 2018, Milwauke, United States, 27/06/18. https://doi.org/10.23919/ACC.2018.8431644

Model Fitting Filter. / Wang, Xiaoxu; Zhang, Jun; Li, Chaofeng et al.
2018 Annual American Control Conference, ACC 2018. Institute of Electrical and Electronics Engineers Inc., 2018. p. 5479-5484 8431644 (Proceedings of the American Control Conference; Vol. 2018-June).

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

TY - GEN

T1 - Model Fitting Filter

AU - Wang, Xiaoxu

AU - Zhang, Jun

AU - Li, Chaofeng

AU - Pan, Quan

AU - Ding, Zhengtao

AU - Yin, Xin

PY - 2018/8/9

Y1 - 2018/8/9

N2 - In this paper, we present a new nonlinear filter based on variational inference, which we have named the model fitting filter (MFF). The core of MFF is to represent or fit the nonlinear measurement likelihood probability with a linear Gaussian probability (LGP) characterized by some parameters that need to be turned. Further, MFF is analytically established on LGP, which is a little similar to the simple linear system filter. Specifically, the nonlinear filtering performance is well improved by utilizing these turned parameters to correct the state estimation. At the same time, MFF does not require the square root decomposition of covariance matrix, so it avoids the inborn defect of numerical instability, which exists more or less in UKF or DDF. The superior performance of MFF is numerically tested by a target tracking problem with nonlinear models.

AB - In this paper, we present a new nonlinear filter based on variational inference, which we have named the model fitting filter (MFF). The core of MFF is to represent or fit the nonlinear measurement likelihood probability with a linear Gaussian probability (LGP) characterized by some parameters that need to be turned. Further, MFF is analytically established on LGP, which is a little similar to the simple linear system filter. Specifically, the nonlinear filtering performance is well improved by utilizing these turned parameters to correct the state estimation. At the same time, MFF does not require the square root decomposition of covariance matrix, so it avoids the inborn defect of numerical instability, which exists more or less in UKF or DDF. The superior performance of MFF is numerically tested by a target tracking problem with nonlinear models.

KW - model fitting filter

KW - Nonlinear estimation

KW - numerical instability

KW - variational inference

UR - http://www.scopus.com/inward/record.url?scp=85052561054&partnerID=8YFLogxK

U2 - 10.23919/ACC.2018.8431644

DO - 10.23919/ACC.2018.8431644

M3 - 会议稿件

AN - SCOPUS:85052561054

SN - 9781538654286

T3 - Proceedings of the American Control Conference

SP - 5479

EP - 5484

BT - 2018 Annual American Control Conference, ACC 2018

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2018 Annual American Control Conference, ACC 2018

Y2 - 27 June 2018 through 29 June 2018

ER -

Model Fitting Filter

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this