首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于FCM的多传感器融合多目标跟踪的数据关联
引用本文:韩红,韩崇昭,朱洪艳,刘允才.基于FCM的多传感器融合多目标跟踪的数据关联[J].系统仿真学报,2004,16(9):2096-2099.
作者姓名:韩红  韩崇昭  朱洪艳  刘允才
作者单位:1. 上海交通大学自动化系,上海,200030;西安交通大学综合自动化研究所,西安,710049
2. 西安交通大学综合自动化研究所,西安,710049
3. 上海交通大学自动化系,上海,200030
基金项目:国家重点基础研究发展规划(973)项目(2001CB309403),863计划项目(20002AA135210),自然科学基金(60172019)
摘    要:数据关联是实现多目标跟踪的核心技术之一,也是实现多传感器信息融合的前提。本文采用改进的模糊c-均值法求解关联概率,并通过在不同的传感器所对应的观测空间上建立多目标运动状态的投影,将单传感器数据关联算法推广到多传感器信息融合系统,从而可在密集杂波环境中实现对多目标的数据关联和精确跟踪。仿真实验结果说明了本文方法的有效性。

关 键 词:模糊聚类  数据关联  多传感器融合  多目标跟踪
文章编号:1004-731X(2004)09-2096-04
修稿时间:2003年8月13日

Multi-sensor Fusion Multi-target Tracking Data Association Based on FCM
HAN Hong,HAN Chong-zhao,ZHU Hong-yan,LIU Yun-cai.Multi-sensor Fusion Multi-target Tracking Data Association Based on FCM[J].Journal of System Simulation,2004,16(9):2096-2099.
Authors:HAN Hong    HAN Chong-zhao  ZHU Hong-yan  LIU Yun-cai
Institution:HAN Hong1,2,HAN Chong-zhao2,ZHU Hong-yan2,LIU Yun-cai1
Abstract:The data association (DA) of track-to-measurement is a kernel problem in the multi-target tracking (MTT) system, and it is also to be the precondition of realizing the multi-sensor information fusion. In this paper, the data association probabilities can be obtained by using the modified fuzzy clustering means (FCMDA) approach, and by means of projecting the target states into the measurement spaces of multi-sensor, the FCMDA algorithm for signal sensor system is extended into the multi-sensor tracking system, and therefore the accurate tracking for multi-targets in heavily cluttered environment can be implemented. The simulation results indicate that the proposed algorithm is very efficient.
Keywords:fuzzy clustering  data association  multi-sensor fusion  multi-target tracking
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号