quotation:[Copy]
[Copy]
【Print page】 【Download 【PDF Full text】 【View/Add Comment】【Download reader】【 Close 】

←Previous page|Page Next →

Back Issue    Advanced search

This Paper:Browse 271   Download 193 本文二维码信息
码上扫一扫!
扩展卡尔曼与改进D-S证据理论结合的多传感器数据融合
高一博,张嘉驰,刘留,章震东,李新星
0
(1.北京交通大学 电子信息工程学院,北京 100044;2.山东警察学院 警务信息学院,济南 250200)
摘要:
提升传感器探测精度对于增强分辨率与优化探测性能具有重要意义,而采用数据融合技术是实现这一目标的有效手段。提出了一种扩展卡尔曼滤波(Extended Kalman Filter,EKF)与改进D-S(Dempste-Shafer)证据理论结合的多传感器数据融合算法,将EKF算法应用于像素级融合,同时引入了D-S证据理论进行决策级融合,通过两层面的融合策略提升数据融合的效能。在融合过程中首先采用EKF对多传感器探测的初始数据进行像素级融合,并比较EKF处理后各结果间的欧氏距离,以D-S证据理论为支撑,选出每一时刻的最佳EKF数据集。通过计算各个融合阶段的均方根误差来评估融合效果。结果表明,该方法能够在保证较高可靠性与较低复杂性的前提下提高融合效果,进而提高传感器的测量精度。
关键词:  多传感器数据融合  扩展卡尔曼滤波  D-S证据理论
DOI:10.20079/j.issn.1001-893x.250125001
基金项目:国家重点研发计划(2023YFB2904801)
A Multi-sensor Data Fusion Combining Extended Kalman Filterand Improved D-S Evidence Theory
GAO Yibo,ZHANG Jiachi,LIU Liu,ZHANG Zhendong,LI Xinxing
(1.School of Electronic and Information Engineering,Beijing Jiaotong University,Beijing 100044,China;2.School of Police Information,Shandong Police College,Jinan 250200,China)
Abstract:
Improving the detection accuracy of multi-sensor is very important to enhance the resolution and optimize the detection performance,and data fusion is an effective method to achieve this goal.A multi-sensor data fusion algorithm combining Extended Kalman Filter(EKF) and improved Dempste-Shafer(D-S) evidence theory is proposed.Specifically,the EKF algorithm is applied to the pixel-level fusion of Kalman Filter data.At the same time,the D-S evidence theory is introduced for decision-level fusion.Through these two levels of fusion strategy,the efficiency of data fusion is improved.In the fusion process,EKF is first used to perform pixel-level fusion of the original data detected by the multi-sensor.The Euclidean distances between different results after EKF processing are compared,which is supported by D-S evidence theory.The optimal EKF dataset is selected for each moment.The root mean square error of each fusion stage is calculated to evaluate the fusion effect.The results show that this method can improve the fusion effect while ensuring high reliability and low complexity,thereby improving the measurement accuracy of the sensor.
Key words:  multi-sensor data fusion  extended Kalman filter  D-S evidence theory