Performance analysis of MEMS gyro and improvement using Kalman filter

Abstract

MEMS gyro has many outstanding advantages like cheap, small, light, less power dissipation, and etc., but its low performance limits its wide application. Based on the self-developed CRG20 MEMS gyro test platform, we experimentally studied Allan variance technique to analysis five common noise of the MEMS gyro. Then AR (1) model is adopted based on time-series data to construct the state equation of the system. In order to improve the accuracy and reduce the noise of the output signal of MEMS gyro, the discrete Kalman filter is introduced and compared with simple filter order filter, Allan variance analysis show that the Kalman filter can effectively restrain the signal's noise and improve the stability and reliability of MEMS gyro through.