摘要
本文提出了一种利用模糊集理论和证据理论的智能传感器数据融合方法,其主要思路为:结合智能传感器的特点首先将每个传感器获取的隶属度函数转化为基本概率指派,再利用改进的组合规则来组合证据,从而得出融合结果。本方法给出了检测数据到基本概率指派的转化方法,还解决了证据组合过程中经常遇到的证据冲突问题。最后借用一个例子阐述了本方法与一般方法的优势,并证明了其应用于实际的有效性。
This paper presents a novel method to intelligent sensor data fusion based on the fuzzy set theory and Dempster-Shafer theory in uncertain environment. The fuzzy membership functions obtained by each sensor have been transformed into basic probability assignment. An improved combination rule is proposed to handle conflict evidence. In this method, the transforming method from monitoring data to basic probability assignment is given and a method to solve the problem of evidence confliction is proposed. For this method, a numerical example is used to show the superiority comparing with the common method and illustrate the validity in practice.
出处
《微计算机信息》
2009年第1期149-150,225,共3页
Control & Automation
关键词
模糊集理论
传感器数据融合
相似性测度
证据理论
fuzzy set theory
sensor data fusion
similarity measure
Dempster-Shafer theory