基于非等间距灰色模型的捷联惯组误差系数建模预测

摘要

Aiming at the problems resulted from small sample and unequal interval testing time, this paper adopts a method of cubic spline function interpolation to adjust the unequal-interval to equal-interval, and the unequal interval grey model is obtained after adopting optimized parameters. The simulation indicates that the method presented in this paper can reflect the changing tendency of the testing data. It is a very useful way to solve the problem of small sample model building for Strapdown Inertial Measurement Unit (SIMU).%捷联惯组历次测试数据由于受到各种因素的影响,数据量少,而且测试时间也是非等间隔的.针对这一特点,提出了捷联惯组误差系数非等间隔灰色建模预测新方法.通过3次样条函数插值,将不等间隔序列调整为等间隔序列；同时将基本预测模型中参数的构造方法加以改进,从而建立非等间隔时序灰色模型.实例分析证明该模型能够准确反映捷联惯组历次测试数据的变化规律,为解决捷联惯组历次测试数据小样本建模问题提供一种可行的途径.