摘要
制导工具误差分离与折合是导弹精度评定中的重要问题。由于工具误差分离存在着很强的复共线性,传统的最小二乘方法与主成分方法不能很好的解决这一问题。提出了利用支持向量机方法获得工具误差系数估计的思想,并将估计结果应用用到弹道误差折合中,与最小二乘和主成分方法相比,支持向量机获得的误差系数估计与真值更加接近,分离残差较小,折合得到的全程弹道遥外差更加接近于真实值。
The separation and conversion of guidance instrumentation systematic error （GISE） play a vital role in the missile precision analysis. Because the separation for GISE is an almost singular regression problem, the classic least square method and principal component method are not satisfactorily for solving the problem. A new method for separating and converting of guidance instrumentation error was proposed which is the support vector machines. The results show that the new method has a higher precision and higher performance than least square method and principal component method.
出处
《系统仿真学报》
EI
CAS
CSCD
北大核心
2007年第10期2177-2179,2182,共4页
Journal of System Simulation
关键词
精度评定
制导工真误差
最小二乘方法
主成分分析
支持向量机
precision assessment
guidance instrumentation error
the least squares method
principal component analysis
support vector machines
