Study on Field Data Fusion and Calibration Techniques of MEMS Array

Due to the poor bias repeatability and large random noise of a micro electro mechanical system (MEMS), the Allan variance was used to analyze the random angle walk of MEMS. The information fusion algorithm of array gyro was designed by using Allan variance identification value and weighted least square method, which could effectively reduce the random angle walk and respond to the true angular rate in real time under both static and dynamic conditions. For the constant drift of MEMS gyro, a two-position calibration scheme was designed combined with the observability of the error of the inertial navigation system, so as to complete system-level calibration of constant drift. Simulation results show that the method proposed in this paper effectively reduces the random angle walk and the constant drift of MEMS gyro and significantly improves the inertial measurement accuracy of MEMS.