北京航空航天大学学报 ›› 2016, Vol. 42 ›› Issue (6): 1107-1115.doi: 10.13700/j.bh.1001-5965.2015.0422

• 论文 • 上一篇    下一篇

基于虚拟观测值的X射线单脉冲星星光组合导航

杨博1, 胡声曼1,2, 孙晖1, 徐帆2   

  1. 1. 北京航空航天大学 宇航学院, 北京 100083;
    2. 北京航天自动控制研究所, 北京 100854
  • 收稿日期:2015-06-24 出版日期:2016-06-20 发布日期:2016-07-06
  • 通讯作者: 杨博,Tel.:15901355288 E-mail:yangbo@buaa.edu.cn E-mail:yangbo@buaa.edu.cn
  • 作者简介:杨博 女,博士,副教授。主要研究方向:航天器自主导航制导、深空探测导航制导和可重复运载器GNC关键技术。Tel.:15901355288 E-mail:yangbo@buaa.edu.cn;胡声曼 女,硕士研究生。主要研究方向:X射线脉冲星自主导航。Tel.:13269830856 E-mail:228737949@qq.com
  • 基金资助:
    国家自然科学基金(51275027)

Integrated single X-ray pulsar and starlight navigation based on virtual observation value

YANG Bo1, HU Shengman1,2, SUN Hui1, XU Fan2   

  1. 1. School of Astronautics, Beijing University of Aeronautics and Astronautics, Beijing 100083, China;
    2. Beijing Aerospace Automatic Control Institute, Beijing 100854, China
  • Received:2015-06-24 Online:2016-06-20 Published:2016-07-06

摘要: 传统的X射线脉冲星导航系统需要同时观测3~4颗脉冲星,有效载荷的质量和功耗极大。因此,单探测器脉冲星导航技术是实现航天器利用X射线脉冲星导航的关键举措。针对单探测器脉冲星导航的可观测性弱和精度低等问题提出了基于虚拟观测值的X射线单脉冲星与星光集中式组合滤波的高精度导航方法,即在X射线脉冲星的长周期内增加与星光同时刻观测的虚拟观测值,以实现高精度的集中式组合滤波算法。同时提出了利用神经网络预测虚拟观测值方法,并与利用动力学递推的方法进行比较,精度可以达到10-7量级。仿真结果表明,该方法可大大提高单探测器的导航的可靠性,补偿由于探测器误差造成的导航误差,导航位置误差为259.79 m,同时有效地减小了导航系统的重量,为X射线脉冲星导航的工程实现提供了参考依据。

关键词: 脉冲星, 单探测器, 组合导航, 神经网络, 虚拟观测值

Abstract: Conventional X-ray pulsar navigation satellite system needs to observe 3 to 4 pulsars at the same time, thus the mass and consumption of the payload will increase. Therefore, the technology using single detector is the key measure to realize X-ray pulsar navigation. In order to solve the problem of poor visibility and low accuracy of X-ray pulsar navigation with single detector, we propose a high accuracy navigation integrated with single X-ray pulsar and starlight navigation based on the centralized Kalman filter using virtual observation value of the pulsar navigation. During the long period of detecting X-ray pulsars, the starlight virtual observations observed at the same time will be added when detecting starlight's observations, so that it can achieve high precision using the centralized Kalman filter within shorter period. Meanwhile, this paper proposes a method using neural network to predict the virtual observation, and compared with the prediction using dynamic method, it can achieve a high order magnitude of 10-7. The results of the simulation show that the method can greatly improve the reliability of the single detector navigation by compensating the navigation error caused by the error of detector the navigation positioning error is 259.79 m, and it can effectively reduce the weight of the navigation system. This can provide reference for the engineering implementation of X-ray pulsar navigation.

Key words: pulsar, single detector, integrated navigation, neural network, virtual observation value

中图分类号: 


版权所有 © 《北京航空航天大学学报》编辑部
通讯地址:北京市海淀区学院路37号 北京航空航天大学学报编辑部 邮编:100191 E-mail:jbuaa@buaa.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发