In submarine surveying and mapping applications, a novel Rauch–Tung–Streibel smoothing (RTSS) scheme based on the factor graph for autonomous underwater vehicles is presented to gain a better offline navigation solution in this paper. The factor graph method is applied to optimally use observation information of multi-sensors with the asynchronous and short-term failure problems to overcome deficiencies of the federal Kalman filter in information fusion processing. Furthermore, the revised RTSS as a post-mission smoothing algorithm is performed by combining the results of the factor graph and one backward data processing through recursively updating the smoothed state and its covariance. From the simulation analysis, it is found that the factor graph mainly owns plug and play capability and contributes to the real-time navigation accuracy over the federal Kalman filtering. The RTSS provides better accuracy and smoothness for the position, velocity, and attitude at the same time compared to the corresponding real-time navigation solution, especially when signals are lost or sensors fail for a short time. With the best of both methods, a novel smoothing scheme combining the factor graph with the RTSS is built. Semi-physical experiment results verify the reliability and effectiveness of the proposed method.
The integration of the strapdown inertial navigation system (SINS) and Doppler velocity log (DVL) has become a basic navigation solution for Autonomous Underwater Vehicles (AUVs). However, DVL cannot obtain the velocity relative to the ground when the distance between the AUV and seabed is over the operating range, which occurs often when AUVs are sailing in the middle layer of the ocean. When the DVL velocity relative to the current is used for an integrated filter, the unknown current velocity is coupled with the measured velocity error, which decreases the positioning accuracy. To address this problem, the effect of unknown coupled current velocity is analyzed from the perspective of filter observability, and an integrated SINS/DVL/virtual velocity navigation method is proposed. The virtual velocity based on the velocity variation extracted from the inertial measurement unit and DVL is constructed and used as an aided measurement for the Kalman filter. With the help of virtual velocity, the current velocity can be easily decoupled from measured SINS velocity error. The results of simulation and experiments demonstrated that the proposed method can effectively improve both the convergence speed and accuracy of velocity error compared with the classical method with SINS/DVL integration and, thus, significantly improve the positioning accuracy.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.