一步数据关联GLMB扩展目标跟踪算法

GLMB extended target tracking based on one-step data association

针对传统的扩展目标跟踪算法在目标近邻场景中由于量测不可分导致跟踪性能下降的问题,提出了一种基于一步数据关联的扩展目标跟踪算法。该算法用乘性噪声模型对目标进行建模,将联合概率数据关联理论中的一步数据关联处理方法与广义标签多伯努利滤波器相结合,实现在量测难以划分情况下的多扩展目标跟踪。仿真实验表明,该算法能够在交叉、近邻场景中实现对目标的有效跟踪,并且在估计精度方面优于传统的基于量测划分的扩展目标跟踪算法。

Due to the inseparability of measurements in neighborhood scenarios,the tracking performance of the traditional extended target tracking algorithm would degrade.In this paper,a new extended target tracking algorithm based on one step data association is proposed to solve the problem.First,the algorithm models the target with a multiplicative noise model.And then,the one step data association method in the Joint Probabilistic Data Association(JPDA)theory is combined with a Generalized Labeled Multi-Bernoulli(GLMB)filter.Simulation results show that the algorithm can track the target in cross and neighborhood scenarios effectively and that it is superior to the traditional extended target tracking algorithms based on measurement partition in estimation accuracy.