Fusion guiding technology solution and algorithm for underwater docking of autonomous underwater vehicles

  Objective  To address the problem of underwater autonomous vehicles (AUVs) docking with parent vessels in recovery operations, a mobile platform-oriented AUV underwater guiding solution is proposed on the basis of multi-sources information fusion, including inertial, acoustic and optical signals.

  Methods  A federated extended Kalman filter integrating multi-sensors information is designed to improve filtering accuracy through decentralized filtering and information fusion. Motion equations are also established by combining the five AUV docking stages, in which the signals detected by the inertial navigation system (INS), acoustic ultra-short baseline (USBL) and optical guiding system are applied separately as inputs of sub-filters, resulting in a fusion guiding algorithm adapted to AUV underwater docking systems.

  Results  Simulation experiments demonstrate that the guiding process based on multi-source information fusion is feasible and possesses robust performance and adequate control and steering accuracy.

  Conclusions  The proposed fusion guiding solution meets the engineering requirements of underwater docking operations. The results of this study can provide technical references for the underwater docking of AUVs.