基于幅度信息的标签多伯努利滤波算法

杂波环境下,现有多目标跟踪滤波器会出现性能衰减。对此,提出了基于幅度信息(amplitude information,AI)的广义标签多伯努利(generalized labeled multi-Bernoulli,GLMB)滤波算法(AI-GLMB)。通常杂波幅度低于目标回波幅度,通过引入幅度信息对目标状态进行扩展,建立幅度似然函数,推导新的更新方程,并给出了算法的序贯蒙特卡罗实现方法。仿真结果表明,AI-GLMB算法能有效适应高杂波环境,同幅度信息辅助的概率假设密度滤波算法、幅度信息势平衡多伯努利滤波算法及传统GLMB滤波算法相比,其跟踪精度更高。

Labeled multi-Bernoulli filter based on amplitude information

For the problem that the existing multiple target tracking filters can cause performance attenuation in clutter environment, an amplitude information generalized labeled multi Bernoulli (AI-GLMB) filter is proposed. In general, the amplitude of clutter is lower than those from target returns. By introducing amplitude information to expand the target states, the amplitude likelihood function is derived. Then the new updating equation is deduced, and the sequential Monte Carlo implementation of the proposed method is given. Simulation results show that the AI-GLMB algorithm can adapt to the clutter environment effectively, and has higher tracking accuracy than the amplitude information assistant probability hypothesis density filter, amplitude information cardinality balanced multi target multi Bernoulli filter and the standard GLMB filter.