摘要
为了削弱陀螺漂移对光纤陀螺寻北精度的影响,在小波阈值消噪的基础上,采用抗差估计处理陀螺信号。首先进行频谱分析,确定相应的多分辨分析尺度及对不同尺度下的高频系数采取的措施。对噪声占主要成份的尺度层其高频系数置零,对噪声与有用信号都存在的尺度层进行小波阈值消噪处理。然后将抗差估计应用于陀螺数据的处理,利用一次抗差估计求得的观测残差再用中位数法求得均方差因子,采用高崩溃污染率的初值辅以IGGⅢ方案迭代解算的混合算法。计算结果表明这种方法能够有效地抵制异常扰动的影响,提高了寻北精(密)度。
A new algorithm based on robust estimation and wavelet threshold de-noising for FOG data is presented to degrade the influence of random drift on the precision of FOG north seeker. First, frequency spectral analysis for FOG data is given to decide the scale of wavelet decomposing and dispose high fre- quency coefficients. The high frequency coefficients of the scale that mostly contains noises will be set zero and those of the scale that contains noises and useful signals will be dealt with the threshold value. Second, robust estimation is applied to FOG data processing, which applies the median of the residuals of the first epoch robust estimation to obtain stable standard deviation factor; adopts the mixed .iterative algorithm based on high breakdown point initial value and assisted with IGGⅢ scheme. The calculations show that the new algorithm can effectively resist external disturbances, and in turn, can improve the precision of FOG north seeker.
出处
《传感技术学报》
CAS
CSCD
北大核心
2009年第11期1596-1601,共6页
Chinese Journal of Sensors and Actuators
基金
军队科研项目资助(K04409)
关键词
光纤陀螺
陀螺漂移
粗差
抗差估计
小波阈值消噪
FOG（fiber optic gyroscope）
gyro drift
outlier
robust estimation
wavelet threshold de-noising
