基于联合概率数据关联和粒子滤波的多目标跟踪算法

Multiple Target Tracking Algorithm Based on Joint Probability Data Association and Particle Filter

数据关联一直是多目标跟踪中的核心问题,是实现多目标有效跟踪的关键。介绍了多目标跟踪的基本原理以及联合概率数据关联的基本原理,并且将粒子滤波引入到联合概率数据关联模型中,提出了联合概率数据关联-粒子滤波算法来实现多目标跟踪。仿真结果表明,此算法可以很好的实现固定数目多目标跟踪。

The data association is the core problem of multiple target tracking, and it is the key point to realize multiple target tracking. The basic theory of multiple target tracking and joint probability data association is introduced. Particle Filter is introduced to the model of joint probability data association. A joint probability data association-particle {ilter algorithm（JPDA-PF） is proposed to realize multiple target tracking. The simulation results indicate that JPDA-PF algorithm can realize multiple target tracking effectively.