一种基于模糊聚类的数据关联新算法#
李良群,易正龙*
(深圳大学 ATR 国防科技重点实验室,广东深圳 518060)
5
10
15
20
25
30
摘要:针对杂波环境下多目标跟踪中的数据关联滤波问题,提出了一种新的直觉模糊联合概
率数据关联滤波方法。与传统的联合概率数据关联滤波器(JPDAF)不同,提出方法能够从
目标观测的不确定信息中提取对目标的隶属信息,以此构建新的直觉模糊隶属度代替 JPDAF
中的关联概率,达到对关联概率的简化快速计算。为了计算直觉模糊隶属度,在模糊 c 均值
聚类的基础上,引入直觉模糊点算子,提出了一种新的直觉模糊聚类方法,同时针对多目标
跟踪的特点,给出了直觉模糊指数的计算方法和一种目标观测的关联不确定性处理方法。仿
真数据和实测数据的实验结果表明,提出方法能够有效对杂波环境的多目标进行关联跟踪,
且性能要好于传统的 JPDAF 和 Fitzgerald’s 方法。
关键词:直觉模糊集;联合概率数据关联滤波器;直觉模糊点算子;模糊聚类
中图分类号:TN953
A Novel Data Association Algorithm Based on the Fuzzy
Clustering
LI Liangqun, YI Zhenglong
(ATR Key Laboratory,Shenzhen University,ShenZhen 518060,)
Abstract: To the data association problem of multi-target tracking in cluttered environment, a new
intuitionistic fuzzy joint probabilistic data association filter was proposed. In the proposed
algorithm, the joint association probabilities in JPDAF are reconstructed by utilizing the
intuitionistic fuzzy membership degree of the measurement belonging to the target. In order to
compute the intuitionistic fuzzy membership degree, a new intuitionistic fuzzy clustering method
is proposed based on the intuitionistic fuzzy point operator, which can extract the useful
information from the unknown information. At the same time, in order to deal with the uncertainty
of the measurements, a new weight assignment is introduced. Moreover, in order to adapt the
multi-target tracking environment, a new intuitionistic index of the new intuitionistic fuzzy set is
defined. Finally, the experiment results based on the simulation data and real-data show the
proposed algorithms have advantages over the JPDAF and Fitzgerald’s JPDAF in terms of
efficiency and putational load.
Key words: Intuitionistic Fuzzy Set; JPDAF; Intuitionistic Fuzzy Point Operator; Fuzzy
Clustering
35
0 引言
杂波环境下多目标跟踪中的数据关联,一直是信息融合领域的难点和热点问题。目前有
效的数据关联方法有:联合概率数据关联(JPDA)、多假设跟踪(MHT)、粒子滤波(PF)等
[1-5]
。在杂波环境下,联合概率数据关联对多个目标的跟踪具有较好的性能,但是,由于在
40
计算目标观测的关联概率时,需要计算可行联合事件的概率,当目标
徐行效声乐心理学与声乐科学体系理论探究 来自淘豆网www.taodocs.com转载请标明出处.