Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review
Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review
}
TY - GEN
T1 - Box-Particle PHD Filter for Multi-Target Tracking
AU - Schikora, Marek
AU - Gning, Amadou
AU - Mihaylova, Lyudmila
AU - Cremers, Daniel
AU - Koch, Wofgang
PY - 2012/7/7
Y1 - 2012/7/7
N2 - This paper develops a novel approach for multitarget tracking, called box-particle probability hypothesis density filter (box-PHD filter). The approach is able to track multiple targets and estimates the unknown number of targets. Furthermore, it is capable to deal with three sources of uncertainty: stochastic,set-theoretic and data association uncertainty. The box-PHD filter reduces the number of particles significantly, which improves the runtime considerably. The small particle number makes this approach attractive for distributed computing. A box-particle is a random sample that occupies a small and controllable rectangular region of non-zero volume. Manipulation of boxes utilizes methods from the field of interval analysis. The theoretical derivation of the box-PHD filter is presented followed by a comparative analysis with a standard sequential Monte Carlo (SMC) version of the PHD filter. To measure the performance objectively three measures are used: inclusion, volume and the optimum subpattern assignment metric. Our studies suggest that the box-PHD filter reaches similar accuracy results, like a SMCPHD filter but with much considerably less computational costs. Furthermore, we can show that in the presence of strongly biased measurement the box-PHD filter even outperforms the classical SMC-PHD filter.
AB - This paper develops a novel approach for multitarget tracking, called box-particle probability hypothesis density filter (box-PHD filter). The approach is able to track multiple targets and estimates the unknown number of targets. Furthermore, it is capable to deal with three sources of uncertainty: stochastic,set-theoretic and data association uncertainty. The box-PHD filter reduces the number of particles significantly, which improves the runtime considerably. The small particle number makes this approach attractive for distributed computing. A box-particle is a random sample that occupies a small and controllable rectangular region of non-zero volume. Manipulation of boxes utilizes methods from the field of interval analysis. The theoretical derivation of the box-PHD filter is presented followed by a comparative analysis with a standard sequential Monte Carlo (SMC) version of the PHD filter. To measure the performance objectively three measures are used: inclusion, volume and the optimum subpattern assignment metric. Our studies suggest that the box-PHD filter reaches similar accuracy results, like a SMCPHD filter but with much considerably less computational costs. Furthermore, we can show that in the presence of strongly biased measurement the box-PHD filter even outperforms the classical SMC-PHD filter.
KW - particle filters
KW - Probability hypothesis density filters
KW - sequential Monte Carlo
KW - Box particle filter
KW - multiple target tracking
KW - on line state estimation
M3 - Conference contribution/Paper
SN - 978-1-4673-0417-7
SP - 106
EP - 113
BT - Information Fusion (FUSION), 2012 15th International Conference on
PB - IEEE
T2 - The 15th International Conference on Information Fusion
Y2 - 9 July 2012 through 12 July 2012
ER -