Characterization of MEMS based Inertial Measurement Unit

Authors:

Abstract
Inertial system consisting of 10 DoF MEMS IMU has been brought to deploy on the stores of high performance fighter aircraft to estimate the separation trajectory of released stores from the aircraft. The IMU calibration has been carried out as it is prerequisite for accurate estimation of stores separation trajectory. This paper presents calibration methodology and test procedure followed to calibrate IMU. The intent is to find out stochastic and deterministic errors in tri-axial rate gyroscopes and tri-axial accelerometers of a MEMS IMU. For this, rate table tests are conducted to calibrate gyroscopes and tumble tests are carried out to calibrate accelerometers. In this paper, the sensor axis misalignment factors in IMU is obtained from gauss-newton optimization algorithm. Thus obtained calibration constants are further used in an EKF based fusion algorithm. A 6-state & a 9-state Extended Kalman filters are presented to estimate Euler angles and position respectively. Effect of calibration on Euler angles and position estimation is clearly brought out.