Home > Research > Publications & Outputs > A convolution particle filtering approach for t...

Electronic data

View graph of relations

A convolution particle filtering approach for tracking elliptical extended objects

Research output: Contribution to conference - Without ISBN/ISSN Conference paper

Published

Standard

A convolution particle filtering approach for tracking elliptical extended objects. / Angelova, Donka; Mihaylova, Lyudmila; Petrov, Nikolay; Gning, Amadou.

2013. 1-6 Paper presented at 16th International Conference on Information Fusion, Istanbul, Turkey.

Research output: Contribution to conference - Without ISBN/ISSN Conference paper

Harvard

Angelova, D, Mihaylova, L, Petrov, N & Gning, A 2013, 'A convolution particle filtering approach for tracking elliptical extended objects', Paper presented at 16th International Conference on Information Fusion, Istanbul, Turkey, 9/07/13 - 12/07/13 pp. 1-6.

APA

Angelova, D., Mihaylova, L., Petrov, N., & Gning, A. (2013). A convolution particle filtering approach for tracking elliptical extended objects. 1-6. Paper presented at 16th International Conference on Information Fusion, Istanbul, Turkey.

Vancouver

Angelova D, Mihaylova L, Petrov N, Gning A. A convolution particle filtering approach for tracking elliptical extended objects. 2013. Paper presented at 16th International Conference on Information Fusion, Istanbul, Turkey.

Author

Angelova, Donka ; Mihaylova, Lyudmila ; Petrov, Nikolay ; Gning, Amadou. / A convolution particle filtering approach for tracking elliptical extended objects. Paper presented at 16th International Conference on Information Fusion, Istanbul, Turkey.6 p.

Bibtex

@conference{6a2b927bf8cb42719a84c13f8e4c5f7f,
title = "A convolution particle filtering approach for tracking elliptical extended objects",
abstract = "This paper proposes a convolution particle filtering approach for extended object tracking. Convolution particle filters (CPFs) are likelihood free filters. They are based on convolution kernel probability density representation. They use kernelsto approximate the likelihood of the observations and represent the likelihood when it is analytically untractable or when the observation noise it too small. Hence, the CPFs represent a sub-family of particle filters with improved efficiency in stateestimation of nonlinear dynamic systems. A CPF is designed and implemented for track maintenance of an object with an elliptical shape. The object kinematics and its extent are estimated in the presence of dense clutter. This nonparametric filter is validated with a Poisson model for the measurements, originating from the target and clutter. Simulation examples illustrate the filter performance. It is shown that the CPF yields correct estimates of the joint probability density function of the state variables and unknown static parameters. The results obtained for the extended objects show that the CPFs provides accurate on-line tracking, with satisfactory estimation of the target shape and volume.",
author = "Donka Angelova and Lyudmila Mihaylova and Nikolay Petrov and Amadou Gning",
year = "2013",
month = jul,
day = "1",
language = "English",
pages = "1--6",
note = "16th International Conference on Information Fusion ; Conference date: 09-07-2013 Through 12-07-2013",

}

RIS

TY - CONF

T1 - A convolution particle filtering approach for tracking elliptical extended objects

AU - Angelova, Donka

AU - Mihaylova, Lyudmila

AU - Petrov, Nikolay

AU - Gning, Amadou

PY - 2013/7/1

Y1 - 2013/7/1

N2 - This paper proposes a convolution particle filtering approach for extended object tracking. Convolution particle filters (CPFs) are likelihood free filters. They are based on convolution kernel probability density representation. They use kernelsto approximate the likelihood of the observations and represent the likelihood when it is analytically untractable or when the observation noise it too small. Hence, the CPFs represent a sub-family of particle filters with improved efficiency in stateestimation of nonlinear dynamic systems. A CPF is designed and implemented for track maintenance of an object with an elliptical shape. The object kinematics and its extent are estimated in the presence of dense clutter. This nonparametric filter is validated with a Poisson model for the measurements, originating from the target and clutter. Simulation examples illustrate the filter performance. It is shown that the CPF yields correct estimates of the joint probability density function of the state variables and unknown static parameters. The results obtained for the extended objects show that the CPFs provides accurate on-line tracking, with satisfactory estimation of the target shape and volume.

AB - This paper proposes a convolution particle filtering approach for extended object tracking. Convolution particle filters (CPFs) are likelihood free filters. They are based on convolution kernel probability density representation. They use kernelsto approximate the likelihood of the observations and represent the likelihood when it is analytically untractable or when the observation noise it too small. Hence, the CPFs represent a sub-family of particle filters with improved efficiency in stateestimation of nonlinear dynamic systems. A CPF is designed and implemented for track maintenance of an object with an elliptical shape. The object kinematics and its extent are estimated in the presence of dense clutter. This nonparametric filter is validated with a Poisson model for the measurements, originating from the target and clutter. Simulation examples illustrate the filter performance. It is shown that the CPF yields correct estimates of the joint probability density function of the state variables and unknown static parameters. The results obtained for the extended objects show that the CPFs provides accurate on-line tracking, with satisfactory estimation of the target shape and volume.

M3 - Conference paper

SP - 1

EP - 6

T2 - 16th International Conference on Information Fusion

Y2 - 9 July 2013 through 12 July 2013

ER -