摘要
为克服原始扩展目标高斯混合概率假设密度滤波算法(Extended Target Gaussian Mixture Probability Hypothesis Density,ET-GM PHD)不能解决机动目标跟踪问题,将交互多模型(Interacting M ultiple M odel,IM M)的思想引入到高斯混合概率假设密度滤波框架下,提出一种交互多模型扩展高斯混合概率假设密度滤波算法(IMM-ET-GMPHD).该算法主要融合了经典的三种运动模型,通过模型的交互实现了对多机动扩展目标的跟踪.此外,为了获取各个机动扩展目标完整航迹,提出一种高斯分量标识方法,使得提出的算法不仅能跟踪多机动扩展目标,还可以有效地估计每个机动扩展目标的航迹.仿真结果表明,本文提出的算法在对复杂环境下多机动扩展目标的跟踪上体现出良好的性能,同时能够有效地管理多机动扩展目标的航迹.
In order to overcome the original extended target Gaussian mixture probability hypothesis density filter algorithm can not solve the problem of tracking maneuvering targets,the Interacting Multiple Model extended target Gaussian mixture probability hypothesis density filter algorithm is proposed based on IMM and ET-GMPHD,the algorithm achieves the multiple maneuvering extended target tracking through the interaction of three classic models.In addition,in order to obtain the trajectories of each target,the Gaussian component labeling method is used to effectively identify each extended target and obtain their trajectories.Simulation results show that the proposed algorithm has better tracking performance for multiple maneuvering extended targets,and can effectively estimate the trajectories of each extended target.
出处
《小型微型计算机系统》
CSCD
北大核心
2018年第2期334-339,共6页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(61305017)资助
江苏省自然科学基金项目(BK20130154)资助
江苏省普通高校研究生科研创新计划项目(KYLX16_0782)资助
关键词
多扩展目标
高斯混合概率假设密度
交互多模型
航迹维持
multiple extended target Gaussian mixture probability density interacting multiple model track maintenance
