一种改进的LAP多目标跟踪数据互联方法

摘要

On the problem of data association efficiency in the tracking using MHT,the process how it works and the reason why it will cause a combined explosive of relevance hypothesis are both analyzed in this paper.Then an improved algorithm based on LAP using likelihood ratio as track value is introduced into the process of MHT data association.The several simulation experiments of motion tracking of cross trajectory are carried out.And the result indicates that,on the premise of tracking accuracy,the improved algorithm can reduce the amount and the complex degree of calculation when establishing and pruning relevance hypothesis.%围绕传统多假设跟踪(MHT)算法跟踪过程中的数据关联效率问题,分析了算法的基本过程,并对造成关联假设组合爆炸的原因进行剖析.在此基础上,提出了一种基于利用似然比航迹得分简化后的线性分配LAP的MHT数据关联方法,并对交叉轨迹运动物体进行了跟踪仿真实验.仿真表明改进后的算法在保证跟踪准确性的前提下,对关联假设和剪枝过程的计算量和复杂度均有一定程度的降低.