Robust self-corrective initial alignment algorithm for strap-down INS

This paper presents a novel robust method of self-corrective Initial Alignment of a stationary SINS¹ with respect to local-level frame (NED-frame) depending only on the output of a strap-down IMU². The proposed method provides a superior performance compared to traditionally-known self-corrective initial alignment methods. The main contribution of this paper is the development of a new algorithm which is based on Robust Kalman Filter theory (see [1]) applied to Multiplicative Attitude Quaternion Kalman Filter MQKF (see [2]). The main advantage of this method is that the resultant filter does not need the models of the inertial sensors errors; but only needs the bounds of their variations.