From the perspective of engineering applications, reducing random drift error of MEMS (microelectromechanical system) gyroscope is one of the main methods to improve gyro performance. In this paper, MEMS gyroscope random drift data are modeled using an auto-regressive-moving-average (ARMA) time series model. Before modeling, software Eviews8.0 is used to perform ADF(Augment Dickey-Fuller) unit root test on drift data to determine whether the data satisfy stationarity. After data preprocessing, the time series model can be identified and the order of ARMA model can be determined by using Akaike Information Criterion (AIC) method. Then the specific method to reduce the drift error using Kalman filtering method based on the time series model is described. The error compensation results of the measured data of a MEMS gyroscope show that the introduced filtering method can effectively suppress its drift error and improve the precision of MEMS gyroscope in practical systems.
Time-serial Modeling and Kalman Filter of MEMS Gyroscope Random drift Compensation
01.08.2018
412871 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Visual approach for MEMS gyroscope drift compensation
IEEE | 2014
|Calibration and compensation periodic error in quartz MEMS gyroscope drift
British Library Online Contents | 2007
|Periodic Error Compensation for Quartz MEMS Gyroscope Drift of INS (P)
British Library Online Contents | 2007
|Novel self-compensation method to lower the temperature drift of a quartz MEMS gyroscope
British Library Online Contents | 2014
|Multiscale fyzzy-adaptive Kalman filtering methods for MEMS gyros random drift
British Library Online Contents | 2009
|