动态总方差及其在陀螺振动信号分析中的应用

Dynamic Total Variance and Its Application in Analysis of Dynamic Characteristics of Gyroscope

动态Allan方差是分析动态环境下陀螺仪随机误差变化规律的一种新方法。针对其运用窗函数截取信号导致方差估计值置信度降低,尤其是长相关时间上估计误差大的问题,提出了动态总方差法。用矩形窗分段截取陀螺仪量测信号,再对截断窗内数据进行倒像映射延拓以增加方差估计的实际自由度,最后计算延拓后样本的Allan方差并提取其噪声系数,将其按时间顺序分别以三维和二维的形式表征出来,以细化和辨识陀螺输出的动态特性。从对半球谐振陀螺线振动试验实测数据处理结果来看,新算法不仅能及时跟踪信号的非平稳变化,而且有效提高了振动信号方差估计值的置信度。

The dynamic Allan variance（DAVAR）is a new method for analyzing random error of gyroscope. Howev-er,it has defects such as poor confidence on the estimate and great estimation error at long-term t-values,due to thereduced amount of data captured by the truncation windows. An improvement dynamic Allan variance is proposed toincrease the confidence in the case of long-term t-values. Firstly,the random error of FOG is truncated with the win-dow function. Then the data in the windows are extended by the total variance method to improve the confidence onthe estimate,At last the Allan variance of extended data is calculated,and the noise coefficients of gyro can also beidentified and extracted at the same time. At last,they are expressed by three or two-dimensional to describe the dy-namic characteristics of gyro. The measured data of linear acceleration test for Hemispherical Resonator Gyro is ana-lyzed with proposed algorithm and dynamic Allan variance. The results shows that proposed algorithm can track thenon-stationary of gyroscope more effectively and obtain a lower estimation error at long-term t-values.