计算机工程与应用 ›› 2019, Vol. 55 ›› Issue (9): 118-123.DOI: 10.3778/j.issn.1002-8331.1801-0400

• 模式识别与人工智能 • 上一篇    下一篇

基于GGIW-CPHD的衍生扩展目标跟踪算法

苗  露,冯新喜,迟珞珈   

  1. 空军工程大学 信息与导航学院,西安 710077
  • 出版日期:2019-05-01 发布日期:2019-04-28

Spawning Expansion Target Tracking Algorithm Based on GGIW-CPHD

MIAO Lu, FENG Xinxi, CHI Luojia   

  1. College of Information and Navigation, Air Force Engineering University, Xi’an 710077, China
  • Online:2019-05-01 Published:2019-04-28

摘要: 针对杂波环境下伽玛高斯逆威舍特混合势概率假设密度(GGIW-CPHD)滤波器难以有效提取衍生扩展目标的问题,提出采用多假设对衍生目标建模跟踪的方法。算法利用随机矩阵模型对扩展目标的形状和尺寸进行建模,并根据多假设模型对衍生事件进行预测,最后通过GGIW混合实现扩展目标运动状态、扩展状态和量测率的联合估计。实验结果表明,与标准GGIW-CPHD滤波算法相比,在含有衍生事件的情景下所提方法实现更好的目标势估计性能且具有较强的适用性。

关键词: GGIW-CPHD滤波器, 衍生目标, 随机矩阵

Abstract: In clutter background, the Gamma Gaussian Inverse Wishart mixture Cardinalized Probability Hypothesis Density(GGIW-CPHD) filter is hard to extract extended target for spawning. A method is proposed to model and track spawning target by using multiple hypothesis structure. The algorithm adopts random matrices models to model the shapes and dimensions of extended target and predicted spawning events according to multi-spawning hypothesis model. It achieves combined estimation of extended target motion state, expansion state and measurement rate via GGIW mixture. Experimental results show that the proposed method can better achieve target cardinality estimation performance and has a stronger applicability in the environment containing spawning events in comparison with standard GGIW-CPHD filter algorithm.

Key words: Gamma Gaussian Inverse Wishart mixture Cardinalized Probability Hypothesis Density(GGIW-CPHD) filter, spawning target, random matrices