中国空间科学技术

• • 上一篇    下一篇

比例UT变换的一种比例因子自适应选取方法

姜伟南;周海银;段晓君;潘晓刚;   

  1. 国防科学技术大学理学院,国防科学技术大学理学院,国防科学技术大学理学院,国防科学技术大学理学院 长沙410073,长沙410073,长沙410073,长沙410073
  • 发布日期:2008-06-25

Self-adaptive Selecting Method for Scaling Factor of Scaled Unscented Transformation

Jiang Weinan Zhou Haiyin Duan Xiaojun Pan Xiaogang(Science College,National University of Defense Technology,Changsha 410073)   

  • Online:2008-06-25
  • Supported by:
    国家自然科学基金项目(60604020);; 航天支撑技术基金项目(2006-HT-GFKD-02)

摘要: 将比例UT(Unscented Transform)变换应用于一般的采样策略可以解决采样的非局部效应问题,提高UKF(Unscented Kalman Filtering)的估计精度,而比例因子是其中的重要参数。文章提出了一种比例因子的自适应选取方法,将自适应的比例UT变换作用于最小偏度单形采样,并把相应的UKF应用于低轨卫星实时定轨中。仿真结果表明,这种自适应选取参数方法稳定可靠,其相应的UKF定轨精度与最佳的固定比例因子UKF的精度相当。

关键词: 无迹卡尔曼滤波, 比例因子, 自适应处理, 单形法, 轨道测定, 卫星

Abstract: By applying the scaled UT(Unscented Transform) to a general sampling strategy,the problem of non-local effects of sampling can be resolved,and the estimating precision of UKF(Unscented Kalman Filter) can be improved as well.The scaling factor is one of the most important parameters,so a self-adaptive selecting method for the scaling factor was proposed.The self-adaptive scaled UT was applied to the scaled minimal skew simplex sampling,and the relevant UKF was applied to the real-time orbit determination of Low-Earth satellites.Simulation results indicate that the method is of high stability,and the relevant precision of the orbit determination approaches to the precision of the UKF with an optimal fixed scaling factor.