Recurrent EKF Based Inertial Aided Magnetometer Online Calibration for Complex Interference Environment

Magnetometer is easily disturbed via external magnetic fields in complex interference environment, which reduces attitude estimation accuracy and seriously affects navigation and positioning accuracy. In this article, we propose an inertial-aided magnetometer online calibration method based on recurrent extended Kalman filter (EKF). First, state equation is constructed in complex interference environment according to relationship between attitude change matrix calculated via gyros and magnetometers output. Then, nonlinear measurement equation is constructed according to magnetometer error calibration model. A recurrent EKF is proposed for solving inertial-aided magnetometer online calibration filtering model. The state estimation accuracy is improved via recurrent filtering measurement updating process to realize high-precision magnetometer online calibration. The simulation and experiment results show that magnetometer calibration accuracy method proposed in this article is superior to the traditional filtering method, and dead reckoning accuracy is improved by 75%.