基于箱粒子滤波的扩展目标伯努利跟踪算法  

Bernoulli filter tracking algorithm for extended object based on box particle

作　　者：孔云波[1] 冯新喜[1] 刘钊[1] 

机构地区：[1]空军工程大学信息与导航学院,陕西西安710077

出　　处：《华中科技大学学报（自然科学版）》2015年第11期63-67,共5页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：国家自然科学基金资助项目(61571458;61403414)

摘　　要：针对复杂环境下,扩展目标跟踪量测存在漏检以及量测不精确、模糊的问题,引入区间分析技术,提出了一种基于箱粒子滤波的扩展目标伯努利滤波算法.算法首先采用伯努利随机有限集和泊松随机有限集分别描述扩展目标的状态和观测,然后将箱粒子滤波应用到扩展目标伯努利滤波算法,推导了适用于不确定量测的扩展目标伪似然函数和状态更新方程,并用伯努利箱粒子滤波递推地对扩展目标的状态进行估计跟踪.仿真实验表明:与扩展目标伯努利粒子滤波算法相比,所提算法在保证扩展目标跟踪性能的同时,有效减少了算法的计算时间.A Bernoulli filter for extended object detection and tracking using interval analysis was pro‐posed for the problem that measurement affected by inaccuracy and vague in the extended target track‐ing .Firstly ,states and observations of extended target were modeled as a Bernoulli random finite set and a Poisson random finite set ,respectively .Then ,the Bernoulli filter was implemented based on box particles ,and the pseudo likelihood function and updated state of extended target applying to in‐terval measurements were derived .Then the state of extended targets was recursively estimated using Bernoulli filter box particles .Simulation results show that compared with the extended target Ber‐noulli particle filter ,the proposed algorithm could lead to significantly lower computational complexity with better tracking performance .

关 键 词：扩展目标 区间分析 伯努利滤波 箱粒子 跟踪算法 

分 类 号：TN953[电子电信—信号与信息处理]