引信多普勒频率的线性稳健拟合

Linear Robust Fitting of Fuze’s Doppler Frequency

摘要:

由于体目标效应影响引信多普勒频率数据经常包含误差(尤其是野值),给飞行试验结果分析带来困难。针对该问题,提出一种用于去除误差的线性稳健拟合方法。首先,建立引信多普勒频率数学模型并对多普勒频率进行线性转换；然后,在分析线性稳健拟合的 M估计法及其权重函数的基础上,给出多普勒频率线性稳健拟合的方法及步骤；最后,对拟合方法及其精度进行仿真验证。仿真结果表明:多普勒频率仅含随机误差时,经一次稳健拟合精度就很高。同时,包含多个野值时,经 2次稳健拟合和 1次野值修正也可达到较高精度。结论是经线性稳健拟合后的引信多普勒频率可用于飞行试验引战配合结果分析以及脱靶量估算。

Abstract:

Fuze’s Doppler frequency often contains errors and outliers because of near-field effect of a target, whichcauses difficulties of testing result analysis. So a linear robust fitting method was put forward to solve this problem. At first,a Doppler frequency mathematical model was established and the linear transform was given. Then, the linear robust fittingmethod and steps were given based on the analysis of M-estimation robust fitting and weight function. At last, the fittingmethod and its accuracy were proved by simulation. The simulation indicated that the accuracy could be very high after on?ly one fitting when Doppler frequency contained random errors, and the accuracy could be high after two fitting and oneoutlier-correction when Doppler frequency contained continuous outliers. The conclusion was that the Doppler frequencyafter linear robust fitting could fulfill the demands of analyzing fuze-warhead coordination and estimating miss distance.