|
|
|
| A New Probabilistic Data Association Filter Based on Probability Theory |
| Liu Zong-xiang; Xie Wei-xin; Huang Jing-xiong |
| College of Information and Engineering, Shenzhen University, Shenzhen 518060, China |
|
|
|
|
Abstract The Probabilistic Data Association Filter (PDAF) and the Joint Probabilistic Data Association Filter (JPDAF) are theoretically analyzed and their shortages in theory are pointed out. Based on the probability theory, a New Probabilistic Data Association Filter (NPDAF) is proposed, in which a measurement may originate from targets or a clutter, but the sum of the probabilities originating from targets and a clutter is equal to 1. Also in the paper, the mathematical model for data association in target tracking and the realization technique for NPDAF are presented. The correlative probabilities between a measurement and targets are first computed in the realization technique, then the gain of a tracking filter is modified using the correlative probability. Simulation results show that the performance of NPDAF is better than that of JPDAF in multiple target tracking.
|
|
Received: 23 June 2008
|
|
|
|
|
|
|
|
|
2009, Vol. 31 
