학술논문
Auxiliary Particle Bernoulli Filter for Target Tracking
Document Type
Article
Author
Source
(2022): 1249-1258.
Subject
Language
Korean
ISSN
15986446
Abstract
Target tracking is a popular topic in various surveillance systems. As a data association free method,the Bernoulli filter can directly estimate target state from plenty of uncertain measurements. However, it is notobvious for existing Bernoulli filters to select proposal distribution with small variance of weights. To address thisproblem, a novel auxiliary particle (AP) Bernoulli filter and its implementation are proposed in this paper. Weemploy the AP method in the Bernoulli filtering framework in order to choose robust particles from a discretedistribution defined by an additional set of weights, which reflect the ability to represent measurements with highprobability. Limitation to the number of particles, the promising particles are used to propagate by extractingindices. On the other hand, the particles without significant contribution to approximation are discarded. In suchcase, the computational complexity of this filter is reduced. With the unscented transform (UT), the dynamics ofmaneuvering target are effectively estimated. The simulation results show advantages in comparison to the standardBernoulli filter for general target tracking.
Target tracking is a popular topic in various surveillance systems. As a data association free method,the Bernoulli filter can directly estimate target state from plenty of uncertain measurements. However, it is notobvious for existing Bernoulli filters to select proposal distribution with small variance of weights. To address thisproblem, a novel auxiliary particle (AP) Bernoulli filter and its implementation are proposed in this paper. Weemploy the AP method in the Bernoulli filtering framework in order to choose robust particles from a discretedistribution defined by an additional set of weights, which reflect the ability to represent measurements with highprobability. Limitation to the number of particles, the promising particles are used to propagate by extractingindices. On the other hand, the particles without significant contribution to approximation are discarded. In suchcase, the computational complexity of this filter is reduced. With the unscented transform (UT), the dynamics ofmaneuvering target are effectively estimated. The simulation results show advantages in comparison to the standardBernoulli filter for general target tracking.
Target tracking is a popular topic in various surveillance systems. As a data association free method,the Bernoulli filter can directly estimate target state from plenty of uncertain measurements. However, it is notobvious for existing Bernoulli filters to select proposal distribution with small variance of weights. To address thisproblem, a novel auxiliary particle (AP) Bernoulli filter and its implementation are proposed in this paper. Weemploy the AP method in the Bernoulli filtering framework in order to choose robust particles from a discretedistribution defined by an additional set of weights, which reflect the ability to represent measurements with highprobability. Limitation to the number of particles, the promising particles are used to propagate by extractingindices. On the other hand, the particles without significant contribution to approximation are discarded. In suchcase, the computational complexity of this filter is reduced. With the unscented transform (UT), the dynamics ofmaneuvering target are effectively estimated. The simulation results show advantages in comparison to the standardBernoulli filter for general target tracking.