A Novel Initial Alignment Method for GPS/INS Integration with Large Initial Heading Error

This paper proposes a novel mechanism for the initial alignment of INS with the aiding of GPS. To use the INS error dynamics based on small attitude errors to achieve the initial alignment of INS, coarse alignment should be carried out prior to fine alignment. However, the general coarse alignment approach by using gyrocompassing theory cannot be used under some circumstances, such as, low-cost systems, and restart of moving systems due to failures. In this paper, a two-phase Kalman Filtering mechanism is proposed for the initial alignment of INS. The first phase is the coarse alignment Kalman Filtering. To solve the problems encountered by the general coarse alignment approach, an INS error dynamics accounting for large initial heading error is developed and used for the coarse alignment, and the corresponding observation equation taking into account of large heading error is also developed. The second phase is the fine alignment Kalman Filtering, where the traditional INS dynamics based on small attitude error is used. Experimental test results have indicated that the proposed alignment approach can achieve the initial alignment more quickly and more accurately comparing with the conventional approaches.