基于带无偏输入的加权融合算法  

Algorithm of weighted fusion based on unbiased input

作　　者：张在利[1] 秦超英[1] 邓奎彪[1] 

机构地区：[1]西北工业大学理学院应用数学系,陕西西安710072

出　　处：《计算机工程与设计》2009年第14期3414-3415,3432,共3页Computer Engineering and Design

摘　　要：给出了带输入估计的卡尔曼滤波算法,比较了带输入估计的卡尔曼滤波算法和标准卡尔曼滤波算法的差别,在此基础上提出了一种最优加权的数据融合方法,分析了最优权值的分配原则,给出了最优加权下的多传感器融合算法。该算法首先计算出各个传感器当前的滤波精度,依据各传感器当前时刻的滤波精度分配权值,同时测量方差的时变特性使得每次测量信息得到充分的利用,通过仿真比较了该方法与平均分配权值方法效果的差别,实验结果表明了该算法的有效性。An unbiased input Kalman filter is advanced, and the difference between unbiased input Kalman filter and standards kalman filter is compared. An optimal weighted method in data fusion is presented. The distribution principle is analyzed for optimal weighted value. The algorithm reasonably distributes weight value according to the filter accuracy renewed of each sensor, meanwhile, the specific property of time-varying in measurement variance makes innovation obtained each time sufficiently utilized. Simulation compared the average value method and the method. And the simulation results show the effective of the new method.

关 键 词：卡尔曼滤波 输入 最优估计 传感器 仿真 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置]