基于PDAF多阵多目标跟踪算法研究

为解决在多目标跟踪过程中，声纳存在交叉运动目标失跟问题．将多阵输出和多目标跟踪相结合，提出了基于概率数据联想滤波(PDAF)器的多阵多目标跟踪算法．计算机模拟结果表明，该算法在平面波模型下，利用两个线阵的输出可以稳定地跟踪两个交叉运动目标．将算法映射到DSP硬件平台，形成了一个多阵多目标跟踪的探测系统，实时处理的结果说明该算法对运动目标的估距误差小于5％

Multi-array multi-target tracking algorithm based on PDAF

In the process of multiple target tracking,sonar usually loses track of cross-moving targets. With the advent of one submarine equiped with several hydrophone arrays,it was possible to combine multiple targets tracking with multiple array outputs. This paper proposes a tracking algorithm based on the probabilistic data association filter (PDAF). Using the model of plane wave incident to two linear arrays,the tracking of two cross-moving targets was studied through computer simulation. The simulation result showed that the algorithm is effective for reliably tracking two crossing targets. Mapping the algorithm onto a DSP hardware platform yielded a detection system with <5% error of range estimation.