基于容积卡尔曼的粒子PHD多目标跟踪算法  

Multitarget tracking with the cubature Kalman particle probability hypothesis density filter

作　　者：王海环[1] 王俊[1] 

机构地区：[1]西安电子科技大学雷达信号处理国家重点实验室,陕西西安710071

出　　处：《系统工程与电子技术》2015年第9期1960-1966,共7页Systems Engineering and Electronics

基　　金：教育部创新团队计划(IRT0954)资助课题

摘　　要：标准粒子概率假设密度(standard particle probability hypothesis density,SP-PHD)滤波在预测粒子状态时没有考虑最新的观测信息,因而存在估计精度较低、粒子退化严重的问题,针对上述问题,提出基于容积卡尔曼的粒子概率假设密度(cubature Kalman particle probability hypothesis density,CP-PHD)滤波算法,该算法基于球面-径向容积数值积分准则,利用容积卡尔曼滤波(cubature Kalman filter,CKF)产生建议密度函数,并对其进行采样得到当前时刻的粒子状态,从而使粒子分布更接近于真实的多目标后验概率密度函数。同时,CP-PHD算法性能不受目标状态维数影响,与无迹卡尔曼粒子概率假设密度(unscented Kalman particle probability hypothesis density,UP-PHD)滤波相比,具有更强适应性和更好的跟踪性能。实验结果表明,CP-PHD算法的跟踪精度优于SP-PHD和UP-PHD。A cubature Kalman particle probability hypothesis density （CP-PHD） filter is proposed to solve the problems of the low state estimation accuracy and the serious particles degradation in standard particle pro- bability hypothesis density （SP-PHD） filter because of unused the most recent observation. CP-PHD uses cubature Kalman filter based on spherical-radial cubature rule to generate the proposal density function and obtains the present particles states by sampling from the proposal density function, so that particle distribution is closer to the real multi-target posterior probability density function. Otherwise, the performance of CPPHD is not af- fected by the dimension of target state, so CP-PHD has stronger adaptive and better tracking performance than unscented Kalman particle probability hypothesis density （UP PHD） filter. Simulation results show that the tracking accuracy of CP-PHD algorithm is superior to SP PHD and UP-PHD.

关 键 词：多目标跟踪 粒子概率假设密度滤波 容积卡尔曼滤波 建议密度函数 

分 类 号：TN953[电子电信—信号与信息处理]