基于模糊推理的多传感器数据融合方法  

Multi-sensor Data Fusion Method Based on Fuzzy Reasoning

作　　者：韩红[1] 韩崇昭[1] 朱洪艳[1] 文戎[1] 

机构地区：[1]西安交通大学电信学院,西安710049

出　　处：《计算机工程与应用》2003年第16期9-10,60,共3页Computer Engineering and Applications

基　　金：国家973重点基础研究发展规划项目资助(编号:2001CB309403)

摘　　要：针对多传感器目标跟踪系统可能出现的传感器漏检现象,提出了一种基于模糊推理的消除漏检现象影响的方法。该算法根据各传感器的量测变化,及时调整参与融合的各传感器量测的数目及其加权系数,保证量测融合值的有效性。通过对两个传感器进行数据融合和目标跟踪的仿真表明,该算法是一种简单有效,有工程应用前景的数据处理算法。该方法同样适用于多于两个传感器的多传感器融合系统。For the sensor miss detection in the multi-sensor data fusion target tracking system,a data fusion algorithm based on fuzzy reasoning is studied in this paper.It provides capability to differentiate between inferior and fine mea-surements of multi-sensor,and can adjust the fused sensor's weight adaptive according to changes in the sensor's mea-surement.The entire data fusion and target tracking system is presented which is based on two sensors system.The simu-lation results indicate that the algorithm is efficient and very potential in practical engineering systems.And it can be extended to an arbitrary number of similar or dissimilar synchronous sensors.

关 键 词：数据融合 模糊推理 相似度 隶属函数 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]