用于机动目标跟踪的联合IMM Viterbi数据关联算法

针对低检测概率、重杂波环境下的机动目标跟踪问题,提出一种联合交互式多模型Viterbi数据关联(C-IMM-VDA)机动目标跟踪算法.该算法利用各模式下的滤波器的量测预测、量测预测协方差及模式概率构造一个统一的综合波门,排除不可能的路径转移,减少到达一个目的节点的源节点的个数.机动目标仿真结果表明,与IMM-VDA(交互式多模型Viterbi数据关联)算法相比,C-IMM-VDA算法在大大减小计算量的同时,降低了航迹失跟率,提高了航迹平均跟踪精度.

A Combined Interactive Multiple Model Viterbi Data Association Algorithm For Manoeuvring Target Tracking

A combined interactive multiple model Viterbi data association (C-IMM-VDA) algorithm is presented for manoeuvring target tracking in the case of low detection probability and heavy clutter. In the C-IMM-VDA, an uniform integrated gate is constructed using measurement prediction, measurement prediction covariance of each sub-filter and mode probability to preclude unlikely path transition between nodes so that the number of possible source nodes can be significantly reduced. Simulation results show the effectiveness of the new algorithm.