当前位置: X-MOL 学术Eur. J. Control › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Multi-target identity management for unknown and time-varying number of targets in clutter
European Journal of Control ( IF 3.4 ) Pub Date : 2021-04-15 , DOI: 10.1016/j.ejcon.2021.03.006
Chiyu Zhang , Dohyeung Kim , Inseok Hwang

A novel multi-target identity management algorithm, called IM-GMPHD (Identity management with Gaussian mixture probability hypothesis density) algorithm, is proposed to simultaneously track an unknown and time-varying number of targets in clutter and manage the targets’ identities effectively and efficiently over time. The targets’ identities are managed by recursively computing the target identity beliefs using the target-measurement correlation. An augmented GMPHD filter is also developed in this paper to calculate the target-measurement correlation. In addition, the target identity beliefs can be updated to a lower uncertainty level whenever there is available target identity information. The proposed multi-target identity management algorithm is demonstrated through various illustrative simulation scenarios.



中文翻译:

多目标身份管理可解决杂乱中未知和随时间变化的目标数量

提出了一种新颖的多目标身份管理算法,称为IM-GMPHD(具有高斯混合概率假设密度的身份管理)算法,可以同时跟踪杂乱中未知且时变的目标数量,并有效地管理目标的身份随着时间的推移。通过使用目标测量相关性递归计算目标身份信念来管理目标身份。本文还开发了一种增强型GMPHD滤波器来计算目标与测量的相关性。另外,只要有可用的目标身份信息,就可以将目标身份信念更新为较低的不确定性级别。通过各种说明性的模拟场景演示了所提出的多目标身份管理算法。

更新日期:2021-05-07
down
wechat
bug