基于模糊聚类的异类多传感器数据关联算法-read.pdfVIP

第 38 卷 　第 4 期 西　安　交　通　大　学　学　报 Vol. 38 　№4 2004 年 4 月 　 Apr. 2004 JOURNAL OF XI′AN J IAOTONG UNIVERSITY 基于模糊聚类的异类多传感器数据关联算法 韩　红 , 韩崇昭 , 朱洪艳 , 左东广 (西安交通大学电子与信息工程学院 , 710049 , 西安) 摘要 : 针对异类传感器观测空间不一致的问题 ,提出了基于模糊聚类的异类多传感器数据关联算法. 该算法 首先通过在不同传感器的观测空间上建立多目标运动状态的投影 ,将多传感器多目标关联问题分解为多个 单传感器多目标的关联问题 ,再对单传感器采用模糊聚类的方法求解关联概率 ,实现了在密集杂波环境中多 目标的数据关联和精确跟踪. 该算法降低了多传感器多目标跟踪的复杂性和计算量 ,有效地解决了异类多传 感器可用公共信息少的问题. 仿真结果表明 ,该算法的跟踪误差要小于传统的联合概率数据关联算法 ,且具 有更优越的跟踪性能. 关键词 : 模糊聚类 ;数据关联 ;多目标跟踪 ( ) 中图分类号: TP274 　文献标识码 : A 　文章编号 : 0253 - 987X 2004 04 - 0388 - 04 Heterogeneous MultiSensor Data Association Algorithm Based on Fuzzy Clustering Han Hong , Han Chongzhao , Zhu Hongyan , Zuo Dongguang ( School of Electronics and Information Engineering , Xi′an Jiaotong University , Xi ′an 710049 , China) Abstract : For the inconsistency problem of heterogeneous sensors’measurement spaces , a new data association (DA) algorithm based on fuzzy clustering algorithm is presented. Firstly , the multisensor multitarget problem is decomposed into the DA problems of several single sensor multitarget by means of projecting the target states into the measurement spaces of the multisensor. Then , the DA probability can be obtained depending on the fuzzy clustering algorithm , and therefore an accurate tracking for multitargets in heavily cluttered environment can be implemented. The presented algo rithm can reduce computing complexity and solve the common informationlacking problem of heterogeneous sensors effi ciently. The simulation results show that the tracking error of propo