A marine SINS transfer alignment method based on group affine of nonlinear state error in local-level frame

Abstract

In a complex and changeable ocean environment, the initial alignment of the ship will be challenging due to the sway induced by the wind and waves, and large initial attitude misalignments cannot satisfy the linear model. Hence, this paper proposes a strapdown inertial navigation system (SINS) transfer alignment method under large misalignment, introduces the Lie group theory, and simultaneously incorporates the attitude and velocity state variables into the special euclidean group (SE (3)), redefining the nonlinear state error to improve the problem of state space inconsistency. The error equation obtained through logarithmic mapping also need satisfies the affine condition. In contrast to the group state model employed in the widely utilisedinertial frame, the local-level frame takes the earth's rotation and Coriolis effect, strong coupling making it challenging to fulfil the group affine property. Therefore, based on the local-level frame, this paper re-derives the error state space model by improving the dynamic equation, thereby realizing the state independence of the group affine model and constructing a new model of transfer alignment under large misalignment. Simulation and experimental test results further verify the effectiveness and superiority of the proposed alignment method under large initial misalignment angles.