中国空间科学技术

• • 上一篇    下一篇

基于补偿最小二乘的航天器轨道确定方法研究

潘晓刚;周海银;   

  1. 国防科学技术大学数学与系统科学系,国防科学技术大学数学与系统科学系 长沙410073,长沙410073
  • 发布日期:2008-04-25

Research on Method of Spacecraft Orbit Determination Based on Penalized Least Squares Estimation

Pan Xiaogang Zhou Haiyin(Department of Mathematic and System Science,National University of Defense Technology,Changsha 410073)   

  • Online:2008-04-25
  • 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: 航天器动力学模型误差是降低轨道确定精度的关键因素之一,文章将航天器动力学模型中的模型误差转化为测量模型的系统误差,建立了测量方程的部分线性模型;并根据补偿最小二乘原理,推导了部分线性模型参数估计方法,并证明了相关性质。仿真试验表明:部分线性模型更接近测量数据的真实表现;基于补偿最小二乘的航天器轨道改进方法能有效逼近航天器状态真值,定轨精度显著优于传统最小二乘估计。

关键词: 部分线性模型, 参数估计, 轨道改进, 补偿最小二乘, 轨道位置估算, 航天器

Abstract: The partial linear observation model was established by transforming the spacecraft dynamic model error into system error of observation model,and the parameter estimation method of partial linear observation model based on Penalized Least Squares Estimation was deduced,and correlative character of the method was proven.Estimation experiment results prove that partial linear model is close to the true characteristic of the measurement data,and the parameter estimation method approaches to the real state of spacecraft,the precision of orbit determination based on penalized least squares estimation is obviously better than the traditional least squares estimation.