Unscented information filtering method for reducing multiple sensor registration error

Y. S. Kim, J. H. Lee, H. M. Do, B. K. Kim, T. Tanikawa, K. Ohba, G. Lee, S. H. Yun

Research output: Contribution to conferencePaper

11 Citations (Scopus)

Abstract

In this paper, new filtering method for sensor registration is provided to estimate and correct error of registration parameters in multiple sensor environments. Sensor registration is based on filtering method to estimate registration parameters in multiple sensor environments. Accuracy of sensor registration can increase performance of data fusion method selected. Due to various error sources, the sensor registration has registration errors recognized as multiple objects even though multiple sensors are tracking one object. In order to estimate the error parameter, new nonlinear information filtering method is developed using minimum mean square error estimation. Instead of linearization of nonlinear function like an extended Kalman filter, information estimation through unscented prediction is used. The proposed method enables to reduce estimation error without a computation of the Jacobian matrix in case that measurement dimension is large. A computer simulation is carried out to evaluate the proposed filtering method with an extended Kalman filter.

Original languageEnglish
Pages326-331
Number of pages6
DOIs
Publication statusPublished - 2008 Dec 1
Event2008 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI - Seoul, Korea, Republic of
Duration: 2008 Aug 202008 Aug 22

Other

Other2008 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI
CountryKorea, Republic of
CitySeoul
Period08/8/2008/8/22

Fingerprint

Information filtering
Sensors
Extended Kalman filters
Error analysis
Jacobian matrices
Data fusion
Linearization
Mean square error
Computer simulation

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Software
  • Computer Science Applications

Cite this

Kim, Y. S., Lee, J. H., Do, H. M., Kim, B. K., Tanikawa, T., Ohba, K., ... Yun, S. H. (2008). Unscented information filtering method for reducing multiple sensor registration error. 326-331. Paper presented at 2008 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI, Seoul, Korea, Republic of. https://doi.org/10.1109/MFI.2008.4648086
Kim, Y. S. ; Lee, J. H. ; Do, H. M. ; Kim, B. K. ; Tanikawa, T. ; Ohba, K. ; Lee, G. ; Yun, S. H. / Unscented information filtering method for reducing multiple sensor registration error. Paper presented at 2008 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI, Seoul, Korea, Republic of.6 p.
@conference{ec025ef950d14a9c9303f4afe6eabeb9,
title = "Unscented information filtering method for reducing multiple sensor registration error",
abstract = "In this paper, new filtering method for sensor registration is provided to estimate and correct error of registration parameters in multiple sensor environments. Sensor registration is based on filtering method to estimate registration parameters in multiple sensor environments. Accuracy of sensor registration can increase performance of data fusion method selected. Due to various error sources, the sensor registration has registration errors recognized as multiple objects even though multiple sensors are tracking one object. In order to estimate the error parameter, new nonlinear information filtering method is developed using minimum mean square error estimation. Instead of linearization of nonlinear function like an extended Kalman filter, information estimation through unscented prediction is used. The proposed method enables to reduce estimation error without a computation of the Jacobian matrix in case that measurement dimension is large. A computer simulation is carried out to evaluate the proposed filtering method with an extended Kalman filter.",
author = "Kim, {Y. S.} and Lee, {J. H.} and Do, {H. M.} and Kim, {B. K.} and T. Tanikawa and K. Ohba and G. Lee and Yun, {S. H.}",
year = "2008",
month = "12",
day = "1",
doi = "10.1109/MFI.2008.4648086",
language = "English",
pages = "326--331",
note = "2008 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI ; Conference date: 20-08-2008 Through 22-08-2008",

}

Kim, YS, Lee, JH, Do, HM, Kim, BK, Tanikawa, T, Ohba, K, Lee, G & Yun, SH 2008, 'Unscented information filtering method for reducing multiple sensor registration error', Paper presented at 2008 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI, Seoul, Korea, Republic of, 08/8/20 - 08/8/22 pp. 326-331. https://doi.org/10.1109/MFI.2008.4648086

Unscented information filtering method for reducing multiple sensor registration error. / Kim, Y. S.; Lee, J. H.; Do, H. M.; Kim, B. K.; Tanikawa, T.; Ohba, K.; Lee, G.; Yun, S. H.

2008. 326-331 Paper presented at 2008 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI, Seoul, Korea, Republic of.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Unscented information filtering method for reducing multiple sensor registration error

AU - Kim, Y. S.

AU - Lee, J. H.

AU - Do, H. M.

AU - Kim, B. K.

AU - Tanikawa, T.

AU - Ohba, K.

AU - Lee, G.

AU - Yun, S. H.

PY - 2008/12/1

Y1 - 2008/12/1

N2 - In this paper, new filtering method for sensor registration is provided to estimate and correct error of registration parameters in multiple sensor environments. Sensor registration is based on filtering method to estimate registration parameters in multiple sensor environments. Accuracy of sensor registration can increase performance of data fusion method selected. Due to various error sources, the sensor registration has registration errors recognized as multiple objects even though multiple sensors are tracking one object. In order to estimate the error parameter, new nonlinear information filtering method is developed using minimum mean square error estimation. Instead of linearization of nonlinear function like an extended Kalman filter, information estimation through unscented prediction is used. The proposed method enables to reduce estimation error without a computation of the Jacobian matrix in case that measurement dimension is large. A computer simulation is carried out to evaluate the proposed filtering method with an extended Kalman filter.

AB - In this paper, new filtering method for sensor registration is provided to estimate and correct error of registration parameters in multiple sensor environments. Sensor registration is based on filtering method to estimate registration parameters in multiple sensor environments. Accuracy of sensor registration can increase performance of data fusion method selected. Due to various error sources, the sensor registration has registration errors recognized as multiple objects even though multiple sensors are tracking one object. In order to estimate the error parameter, new nonlinear information filtering method is developed using minimum mean square error estimation. Instead of linearization of nonlinear function like an extended Kalman filter, information estimation through unscented prediction is used. The proposed method enables to reduce estimation error without a computation of the Jacobian matrix in case that measurement dimension is large. A computer simulation is carried out to evaluate the proposed filtering method with an extended Kalman filter.

UR - http://www.scopus.com/inward/record.url?scp=67650536487&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67650536487&partnerID=8YFLogxK

U2 - 10.1109/MFI.2008.4648086

DO - 10.1109/MFI.2008.4648086

M3 - Paper

AN - SCOPUS:67650536487

SP - 326

EP - 331

ER -

Kim YS, Lee JH, Do HM, Kim BK, Tanikawa T, Ohba K et al. Unscented information filtering method for reducing multiple sensor registration error. 2008. Paper presented at 2008 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI, Seoul, Korea, Republic of. https://doi.org/10.1109/MFI.2008.4648086