The EKF-based SINS/DVL integrated navigation for AUV on lie group under hovring condition

When autonomous underwater vehicles (AUVs) perform tasks under hovering conditions, the strapdown inertial navigation system (SINS) and doppler velocity log (DVL) integrated navigation system encounters significant observability constraints. It is not feasible to enhance observability through motion excitation or the addition of external sensors, leading to partial unobservability of the system's state variables. This results in inconsistencies during the extended kalman filter (EKF) process, causing filter divergence. To address this issue, this paper proposes a SINS/DVL integrated navigation algorithm based on Lie group, redefining the system's nonlinear state errors and measurement models. To circumvent the limitations of conventional inertial frames, a new auxiliary vector is introduced in the local-level frame to achieve affine improvement, and a lie group-based EKF based on group affine is obtained. This approach not only effectively reduces inconsistencies caused by unobservable state variables but also enhances the stability of the filtering system under hovering conditions. Finally, theoretical analysis and simulation results demonstrate that the proposed algorithm offers certain advantages in terms of heading attitude estimation and positioning accuracy.