针对制导工具系统误差分离中环境函数矩阵严重病态,最小二乘估计和主成分估计结果不准的问题,提出了采用岭型主成分估计改善估值结果的方法,分析了该方法的估计结果具有更小的均方误差和更高估计精度。计算机仿真结果证明,在29项制导工具系统误差分离中,岭型主成分估计能在一定程度上克服环境函数矩阵病态的影响,岭型主成分估计分别比最小二乘估计和主成分估计的估准项数多8项和2项。

environmental function matrix is morbidity in guidance instrument systematic error estimation , so least squares estimation and principal components estimation is unfaithful. Combining ridge and principal components estimate can lighten matrix morbidity and improve result precision, it has higher estimation accuracy, The computer simulation show that combining ridge and principal components estimate can lighten matrix morbidity in guidance instrument systematic error estimation, it have more good Estimate result than least squares estimation and principal components estimation.