摘要
在多目标跟踪过程中,针对概率假设密度滤波器难以正确估计目标个数和目标状态问题,提出一种新的基于边缘粒子滤波的改进算法。算法运用Rao一Blackwellized思想,将目标状态分解为线性和非线性模型的结构,采用RBPF滤波方法预测与估计概率假设密度滤波器中目标的非线性状态,使用卡尔曼滤波方法对线性状态进行预测与估计,以更好地提高目标状态估计精度,降低了计算的复杂度。文章最后进行了仿真实验验证,与现有算法相比较,提出的算法能够更加准确地估计出目标个数和目标状态,具有较好的跟踪性能。
In the process of multi-target tracking,a new algorithm based on edge particle filter is proposed to solve the prob-lem that the probability density filter does not estimate the number of targets and the state of target correctly. Application of Rao-Blackwellized algorithm,the target state is decomposed into linear and non-linear model structure. And RBPF filter prediction isused to predict and estimatethe non-linear stateof target,and Calman filtering method is used to predict and estimate the linearstate,in order to improve the estimation accuracy of target state,and the complexity of calculation is reduced. Finally,the simula-tion experiments are carried out to verify the proposed algorithm. Compared with the existing algorithms,the proposed algorithm canestimate the number of targets and thestate of target more accurately,and has better tracking performance.
作者
石治国
吴铭
郝云鹏
施冬磊
SHI Zhiguo;WU Ming;HAO Yunpeng;SHI Donglei(No. 92853 Troops of PLA,Huludao 125106)
出处
《计算机与数字工程》
2019年第2期344-348,共5页
Computer & Digital Engineering
关键词
粒子滤波
概率假设密度滤波
多目标跟踪
particle filter
probability hypothesis density filter
multitarget tracking