传感技术学报2017，Vol.30Issue(11)：1666-1670,5.DOI:10.3969/j.issn.1004-1699.2017.11.009
基于自适应Kalman滤波的MEMS陀螺随机误差分析
Random Error Analysis of MEMS Gyroscope Based on Adaptive Kalman Filter
王辛望 1沈小林 1刘新生2
作者信息
- 1. 中北大学计算机与控制工程学院,太原030051
- 2. 江苏曙光光电有限公司,江苏 扬州225009
- 折叠
摘要
Abstract
In order to improve performance of a certain type MEMS gyroscope,based on the principles of time series analysis,ARMA model is established and ARMA(2,1)is used to establish MEMS gyroscope random error model. The Kalman filter is designed and the result of static test and the constant rate test show that under the classic Kal-man filter,the mean and mean square deviation of the MEMS gyroscope random error is reduced by 32.62% and 66.31% in the static test;the mean is much smaller and the mean square deviation is decreased by an order of mag-nitude in the constant rate test. Based on the fact that the classic Kalman filter can not adapt to the vibration test of large amplitude,a new adaptive Kalman filter is proposed in this paper by looking for the adaptive calibration factors to deal with the problem of the divergence in the classic Kalman filter. The results of vibration test show that the mean and the mean square deviation after filtering is reduced by 8.25% and 8.36% when the amplitude is 100°.关键词
MEMS陀螺/随机误差/自适应Kalman滤波/时间序列分析/自回归滑动平均/Allan方差Key words
MEMS gyroscope/random error/adaptive Kalman filter/time series analysis/Auto-Regressive and Moving Average/Allan variance分类
航空航天引用本文复制引用
王辛望,沈小林,刘新生..基于自适应Kalman滤波的MEMS陀螺随机误差分析[J].传感技术学报,2017,30(11):1666-1670,5.
