摘要
针对制导误差分离模型中环境矩阵S存在严重病态性,从而影响分离结果精度问题,提出了一种基于动力系统求解的制导误差分离方法。该方法从分析线性迭代求解方法入手,将具有病态特性的线性方程组求解问题转化为对相应刚性动力系统的求解问题。这里给出了该方法收敛性及其他特性的证明。为了验证该方法效果,在遥外测视速度误差分别为0.01m/s、0.02m/s以及0.03 m/s的条件下,选用PB(Primary Bayesian,主成分贝叶斯)估计方法与其进行比较,数值结果表明,该方法可有效地降低环境矩阵病态性对误差分离结果的影响,且分离结果的稳健性和精度都优于PB估计方法得到的结果。
Environmental matrix S in the guidance instrument systematic error model is usually seriously ill-conditioned and has a strong impact on accuracy of the result of error separation. In this paper 〉 a new algorithm for guidance instrument er- ror separation is presented by analysis of some numerical iterative solutions. The new algorithm converts solving of Hl-condi- tioned algebraic equations to solving of corresponding stiff dynamic system. The convergence and other characteri new method are proved. To verify effectiveness of the method, Primary Bayesian （PB） method is chosen to compare with it. On the condition of error 0. 01 m/s, 0. 02 m/s, and 0. 03 m/s, numerical experiments were carried out. The results show that the new method is more effective for guidance instrument error separation.
出处
《飞行器测控学报》
CSCD
2017年第1期19-24,共6页
Journal of Spacecraft TT&C Technology
关键词
制导误差
刚性动力系统
病态
误差分离
guidance error
stiff dynamic system
ill-conditioned
error separation
