Computer Engineering and Applications ›› 2012, Vol. 48 ›› Issue (3): 111-113.

• 数据库、信号与信息处理 • Previous Articles     Next Articles

Joint probability data association algorithm with fusing multi-feature information

GAO Qian1, ZOU Hailin1, LIU Chanjuan1,2, ZHOU Li1   

  1. 1.School of Information & Electrical Engineering, Ludong University, Yantai, Shandong 264025, China
    2.School of Mechanical Electronic & Information Engineering, China University of Mining & Technology, Beijing 100083, China
  • Received:1900-01-01 Revised:1900-01-01 Online:2012-01-21 Published:2012-01-21

融合多特征信息的联合概率数据关联算法

高 倩1,邹海林1,柳婵娟1,2,周 莉1   

  1. 1.鲁东大学 信息与电气工程学院,山东 烟台 264025
    2.中国矿业大学(北京) 机电与信息工程学院,北京 100083

Abstract: Joint Probability Data Association(JPDA) algorithm only uses the newest status measurements. As for this shortcoming, this paper proposes an improved JPDA algorithm which fuses multiple feature information. The new algorithm calculates the association matrix between each feature information and targets. According to D-S theory of evidence, this paper fuses status measurements with the other feature information to get fused association probability. Then the fused association probability will be used to modify the original association probability gotten by using JPDA algorithm. And the modified association probability will be used to update the state of targets. Compared to JPDA algorithm, simulations show that the tracking error of the new algorithm can be decreased from 27 to 60 percent.

Key words: joint data association, D-S theory of evidence, feature information, course information, data fusion

摘要: 针对传统联合概率数据关联(JPDA)算法仅利用传感器状态测量信息的不足,提出了一种融合目标多种特征信息的改进JPDA算法。该算法首先根据各种特征信息和目标之间关联度的定义,计算出各种特征信息的测量值与目标之间的关联度矩阵,然后利用D-S证据理论融合状态测量和多种特征信息,最后用融合后的关联概率修正JPDA算法得到的关联概率,以此对目标的状态进行更新。仿真实验表明,与原有的JPDA算法相比,所提改进算法的跟踪误差可降低约27至60个百分点。

关键词: 联合概率数据关联, D-S证据理论, 特征信息, 航向信息, 信息融合