Abstract: The Particle Probability Hypothesis Density filter（P-PHD） has emerged as an effective way to solve the multi-target tracking problems. Multi-target states are expressed in a series of particles with weights. It is of importance to estimate the target state from those particles in the multi-target tracking procedure. A new method to extract target states without the need of clustering is proposed. The updating step of the P-PHD is decomposed. The observation categories coming of the real-target are selected and the chosen observation categories are assigned with the same number of new particles with new weights respectively. The target state is extracted from those brand new particle clouds directly and there is no need to execute the clustering and peak extraction operation. The simulation results demonstrate that the proposed algorithm has a better performance than the k-means method.

Key words: Probability Hypothesis Density（PHD） particle filter, multi-target tracking, state extraction

摘要： 概率假设密度粒子滤波（P-PHD）以粒子集形式反映目标的状态信息，是一种有效的多目标跟踪方法，其关键步骤是从粒子集中准确提取多目标状态信息。提出一种免聚类概率假设密度粒子滤波多目标状态提取方法，通过分解P-PHD迭代更新过程，筛选疑似真实目标量测类别，并重新分配粒子集，根据新粒子集直接提取目标状态，可避免粒子中心聚类和粒子峰值提取过程。仿真结果表明该方法具有较高状态提取精度。

关键词: 概率假设密度粒子滤波, 多目标跟踪, 状态提取