Development of real-time heading estimation algorithm using magnetometer/IMU

This paper presents a real-time heading estimation algorithm using IMU and strapdown magnetometer without any other external heading reference system. To calibrate the output deviation of magnetometer, sensor errors caused by hard, soft iron effect and initial heading of strapdown magnetometers are considered. First, for the estimation of heading angle, system and measurement model is derived, which reveals to be nonlinear. Then particle filter and extended Kalman filter is introduced for the performance comparison. The proposed algorithm for the integration of IMU and magnetometer is verified via numerical simulation in Matlab. Simulation result demonstrates accurate heading estimation error less than 1 degree for both algorithms when there exist small initial heading error and hard iron effect, yet particle filter provides more robust and precise result than the extended Kalman filter in case the initial heading error and biases are large.