首页 | 本学科首页   官方微博 | 高级检索  
     

改进UKF在X射线脉冲星导航中的应用研究
引用本文:邓新坪,郑建华,高东. 改进UKF在X射线脉冲星导航中的应用研究[J]. 北京理工大学学报, 2012, 32(1): 51-55
作者姓名:邓新坪  郑建华  高东
作者单位:中国科学院国家空间科学中心,北京 100190;中国科学院研究生院,北京100190;中国科学院国家空间科学中心,北京,100190
基金项目:国家部委预研项目(XDA04074300)
摘    要:为了提高标准Unscented卡尔曼滤波(UKF)计算效率,本文对标准UKF采样点计算方法进行改进,使状态更新过程中的多次积分过程简化为一次积分过程,并证明了标准UKF和改进UKF的等效性.计算机仿真结果表明,在计算精度一致(位置确定精度约为70m,速度确定精度约为0.005m/s)的情况下,改进UKF的计算效率是标准UKF的5倍.

关 键 词:Unscented卡尔曼滤波  改进Unscented卡尔曼滤波  计算效率
收稿时间:2011-01-18

Application Research of Modified Unscented Kalman Filter in X-Ray Pulsar Navigation
DENG Xin-ping,ZHENG Jian-hua and GAO Dong. Application Research of Modified Unscented Kalman Filter in X-Ray Pulsar Navigation[J]. Journal of Beijing Institute of Technology(Natural Science Edition), 2012, 32(1): 51-55
Authors:DENG Xin-ping  ZHENG Jian-hua  GAO Dong
Affiliation:National Space Science Center, Chinese Academy of Sciences, Beijing 100190, China; Graduate University of Chinese Academy of Sciences, Beijing 100190, China;National Space Science Center, Chinese Academy of Sciences, Beijing 100190, China;National Space Science Center, Chinese Academy of Sciences, Beijing 100190, China
Abstract:It is known that the Unscented Kalman filter (UKF) is applied in X-ray pulsar navigation because of the high nonlinearity of its state equation and measurement equation. In order to improve the computational efficiency of standard UKF, the method to get sample points in UKF is modified in this paper. As a result, the state update could be done by integration needed only one time. Moreover, the equivalency of standard UKF and modified UKF is proved in the work. The simulation results demonstrate that the computational efficiency of modified UKF could be 5 times higher than that of standard UKF with the same condition of computation precision:the position determination precision is about 70 meters and the velocity determination precision is about 0.005 meters per second.
Keywords:Unscented Kalman filter  modified Unscented Kalman filter  computational efficiency
本文献已被 CNKI 万方数据 等数据库收录!
点击此处可从《北京理工大学学报》浏览原始摘要信息
点击此处可从《北京理工大学学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号