MEMS陀螺仪随机误差的Allan分析

为了减少MEMS陀螺仪的误差并提高其精度,需要对陀螺仪误差进行估算与补偿,因而建立陀螺仪的随机误差模型。在陀螺仪随机误差模型分析方法中,有功率谱密度分析、时序ARMA模型及Allan方差分析。Allan方差分析是在时域上对信号频率稳定性进行分析的一种通用方法。通过分析Allan方差,可以分辨出存在于MEMS陀螺中的各种类型噪声。文中用Allan方差对MEMS陀螺仪进行具体分析,得到存在于陀螺仪信号中的各误差源。实验结果表明,Allan方差分析是建立陀螺仪随机误差模型的一种很实用的方法。

Allan Analysis Method for MEMS Gyroscope Random Error

In order to reduce the MEMS gyroscope error and increase the precision,it needs to carry on the estimation and compensation for the gyroscope error.Thus the gyroscope random error model is established.There are many analysis methods for the gyroscope random error model,such as the power spectral density(PSD) analysis method,the succession ARMA model and Allan variance analysis method.The Allan variance analysis method is a general method for the stability analysis of signal's frequency in the time domain.By analyzing the Allan variance, the type of the noise existing in the MEMS gyroscope can be distinguished.This paper accomplishes the concrete analysis with the Allan variance to the MEMS gyroscope,and obtains various error sources existing in the gyroscope signal.The test result indicates that the Allan variance analysis method is practical for the establishment of the gyroscope random error model.