A HEADING CONSTRAINT CALIBRATION METHOD FOR LOW-END INERTIAL MEASUREMENT UNITS
Keywords: IMU calibration Method, MEMS Inertial Sensors, Kalman Filtering, Pseudo-observation
Abstract. The calibration result of the IMU has a strong impact on the accuracy of inertial navigation and its integration with other navigation techniques. Thus, how to efficiently obtain high-precision IMU calibration results is an important research problem for localization and motion tracking with consumer devices. To solve this problem, this paper proposes a handheld calibration method. Similar to our previous work, the pseudo observation is used to replace the measurement equation of the Kalman filter in the GNSS/INS loosely-coupled navigation algorithm. Compared to the existing online calibration algorithm, a more convenient data acquisition method is used, and the heading constraint information is added to assist in obtaining the calibration results of the IMU. To verify the proposed algorithm, a simulator is used to generate the heading updates with various precisions. The proposed algorithm shows the potential to estimate the vertical gyro bias, which does not converge in the existing calibration method, within around 0.5s when the accuracy of the heading’s random error is 5 degrees. When the heading random error is 60 degrees, the vertical gyro bias can converge in about 6 seconds after rotation with the standard deviation of 121.7765 deg/h.