组合导航系统离线信息融合算法  

Algorithm of post-information fusion on integrated navigation system

作　　者：邱吉冰[1] 蒋志翔[1] 

机构地区：[1]中国航天科工集团第二研究院706所研发中心,北京100854

出　　处：《计算机工程与设计》2008年第9期2346-2349,共4页Computer Engineering and Design

摘　　要：获得高精度的载体实际飞行轨迹,是组合导航系统性能评测的重要手段。从组合导航系统研究出发,针对惯性导航系统、姿态(非姿态)型GPS和大气数据计算机多传感器组合导航方式,实现多传感器信息离线融合算法。同类型导航传感器数据采用加权最小二乘算法,并利用开环卡尔曼滤波实现组合导航的信息融合,在此基础上再进行最优固定区间平滑滤波。仿真实验结果表明,经过上述3步信息融合之后,得到的导航结果精度不仅高于任一个传感器精度,而且高于单独采用卡尔曼滤波的导航参数精度。In order to obtain the high accurate actual flight path of carrier, under the research of integrated navigation system, in view of multi-sensor integrated navigation system of inertial navigation system, attitude （non-attitude） GPS and air data computer, the algorithm of post-information fusion on multi-sensor is realized. The weighting least-square technique is used for similar navigation sensor data, and on the foundation of open-loop Kalman filtering, the best fixed-interval optimal smoothing filtering is used for combined navigation information fusion. The experimental result indicated, after the above three steps of information fusion, the navigation result precision is higher than not only any sensor precision, but also Kalman filter used alone.

关 键 词：组合导航 全球定位系统 卡尔曼滤波 固定区间平滑 最小二乘 

分 类 号：TP39[自动化与计算机技术—计算机应用技术] U666.1[自动化与计算机技术—计算机科学与技术]