多传感器信息融合的结构及其算法  

Structure and Algorithm in Multi-sensor Data Fusion

作　　者：郭雁文[1] 兰艳亭[2] 

机构地区：[1]中北大学光电仪器厂,山西太原030051 [2]中北大学信息与通信工程学院,山西太原030051

出　　处：《太原师范学院学报（自然科学版）》2008年第1期67-69,共3页Journal of Taiyuan Normal University:Natural Science Edition

摘　　要：文章介绍了多传感器信息融合的四种结构及其算法.并阐述了各自的优缺点,集中式最大的优点是信息损失最小,但计算负担重;分散式结构,系统的可靠性和容错性高,计算和通信负担比集中式要轻;分级融合结构,其计算和通信负担介于集中式结构和分散式结构之间.There introduces the four structures of data fusion of the multi-sensor and its algorithm.Meanwhile it expounds the advantages and disadvantages of these four structures respectively.The most important advantage of the centralized structure is its less-lose of information,but has much calculation burden.Distributed structure is famous for its systematic reliability,high capacity for errors,and less calculation and communication burden than the centralized ones;the calculating and communication burden of the classified fusion structure is between centralized ones and the distributed ones.

关 键 词：多传感器 信息融合 KALMAN滤波 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置]