修正概率数据关联算法的研究

A Modified Practical Multitarget Data Association Algorithm

根据“全邻”数据关联算法的基本思想,采用关联城内回波似然函数的归一化处理,考虑关联城内所有回波的信息,对Fitzgerald提出的简化概率数据关联(SPDA)算法进行修正,得到修正概率数据关联(MSPDA)算法.理论分析和Monto Carlo仿真表明了该算法的有效性.为了能应用于多机动目标的跟踪.采用一种有效的交互式多模型自适应机动目标跟踪算法与MSPDA算法结合,对多个机动目标的交叉、编队进行仿真,得到了较好的结果.

SPDA (Simplified Probabilistic Data Association) algorithm, proposed by Fitzgerald [4], requires much less computations as compared with JPDA (Joint Probabilistic Data Association) at the sacrifice of disregarding the effects of all returns in validation region. This paper discusses PDA, JPDA and SPDA in detail and proposes a modified probabilistic data association algorithm. The main contributions of this paper are as follows: Using normalized processing of likelihood functions of the returns in validation region, we derive the formula of an amendment coefficient r_1^(k). By r_1^(k), we can consider the effects of all returns in validation region at the sacrifice of just a small increase in amount of computations as compared with SPDA. We use a new interacting mult-model adaptive algorithm to further improve the tracking performance of our proposed algorithm in the case of multiple maneuvering and non-maneuvering targets. Theoretical analysis and Monte Carlo simulations show that the method of this paper is efficient.