基于回归分析的测角系统误差修正方法

Error Correction Method of Angle Measuring System Based on Regression Analysis

回归分析是一种常用的数据拟合方法,可以局部消除误差,并对数据的变化趋势进行分析预测。文章借鉴同轴跟踪处理系统误差的方法,分析了测角系统误差的主要来源,及其与方位、俯仰角之间的变化关系。在大量分析典型海域实测数据的基础上,根据测量数据的特点,采用最小二乘法求出回归系数,从而得出了角度与时间的实时变化关系,以实时修正测角系统误差。修正后的方位和俯仰角度与时间呈线性变化关系,能有效地对角度数据进行外推、预测,减小了系统误差,大幅提高了测角精度,对于提高大型测控系统的实时交汇精度有一定的实用价值。

Regression analysis is a frequently-used data fitting method,which can eliminate errors locally and predict the trend of data change. In this paper,the main sources of angle measurement system errors and their relationship with azimuth and elevation are analyzed by referring to the method of coaxial tracking to deal with system errors. On the basis of analyzing a large number of measured data in typical sea areas,according to the characteristics of measured data,the regression coefficient is calculated by least square method,and real-time relationship between angle and time is obtained to correct the systematic error of angle measurement in real time. The corrected azimuth and elevation have a linear relationship with time,which can effectively extrapolate and predict the angle data,reduce the system error and greatly improve the accuracy of angle measurement. It has a certain practical value for improving the real-time intersection accuracy of large-scale TT&C system.