舰载红外警戒系统的单站多目标数据关联与滤波算法

Data association and tracking algorithm for a shipborne infrared surveillance system

针对舰载红外警戒系统和所观测目标的特点,本文提出一种单站被动式跟踪系统的关联滤波算法.首先选取极坐标系建模使系统方程得到简化,然后基于启发式规则进行航迹起始,并给出纯角度域下的联合概率数据关联算法(JPDA),接着分析变量在角度域的变化规律,结合实际对算法进行修正从而完成交互多模型(IMM)的跟踪滤波.不同的仿真场景被给出用于验证该算法的性能,蒙特卡罗仿真结果表明该算法的有效性.

Based on the characteristics of the infrared sensor and targets, we propose a tracking algorithm for a single passive sensor system. The polar coordinates are adopted in modeling the system for simplicity. A method based on heuristic rules is utilized to initiate the track, and the joint probability data association （JPDA） in the bearing-only field is presented. Based on the variation rule of the angular vectors, we make a modification to the algorithm and implement the interacting multiple model （IMM） tracking filter. Several simulating scenarios are provided to verify the performance of the algorithm, and the simulation results demonstrate the validity of the new method.