GNSS/SINS/DVL Integrated Navigation Algorithm Based on State Transformation Extend Kalman Filtering

Unmanned Underwater Vehicle (UUV) typically employs the Doppler velocity logger/Strapdown Inertial Navigation System (DVL/SINS) integrated navigation mode for underwater operations, yet it need to surface periodically to receive Global Navigation Satellite System (GNSS) signals to correct position errors accumulated in DVL/SINS integration. To tackle the coordinate system inconsistency issue in the error definition of the SINS, this paper uniformly projects velocity errors onto the true navigation coordinate system and thereby proposes an integrated navigation algorithm based on the State Transformation Extended Kalman Filter (STEKF). Based on this newly defined velocity error, more rigorous error equations for SINS attitude, velocity, and position are derived, and an improved state-space model as well as a loose integration measurement model for GNSS/SINS/DVL integrated navigation are constructed. Simulation results demonstrate that, compared with the Extended Kalman Filter (EKF) algorithm, the proposed algorithm shortens the convergence time of the heading misalignment angle from 180 s to 80 s, reduces the position error at the navigation endpoint by approximately 50%, and maintains good robustness even when the initial heading misalignment angle reaches 9°. Field test verifications show that the algorithm decreases the Root Mean Square Error (RMSE) of the northward position error by 15%, the RMSE of the eastward misalignment angle by 51%, and the RMSE of the heading misalignment angle by 20%. The research results confirm that the proposed algorithm can effectively mitigate the impact of coordinate system inconsistency in traditional error models and significantly enhance the positioning accuracy and stability of the GNSS/SINS/DVL integrated navigation system.