A robust and fast self-alignment method for strapdown inertial navigation system in rough sea conditions

Abstract

In this paper, a robust and fast self-alignment method under an undesirable non-periodic unidirectional vessel motion is proposed. The proposed method is based on the attitude determination technique which transforms the alignment procedure into an attitude determination problem by creating infinite vector observations. The method implements the QUaternion ESTimator (QUEST) attitude determination technique which reduces the computational complexity in practical implementations and accomplishes the alignment task more robustly and accurately compared with the Kalman filter based alignment methods in a specific time interval. Simulation and also experimental results show that the proposed technique has more accurate and faster responses compared to the conventional methods based on the Kalman filter. • Introducing a vessel motion derived from experience during self-alignment of SINS. • Overcoming the challenge of self-alignment by creating vector observations. • Low computational complexity for attitude determination by using vector observations. • The robustness of the proposed method against the disturbing forces. • Obtaining the vector observations for compensation of the Lever Arm effect.