Research On AUV Docking Navigation Based On Information Fusion With RSS And Monocular Vision

The undersea observation network is a powerful platform for humans to monitor and explore oceans,and Autonomous Underwater Vehicles(AUVs)are a necessary part of the network,which provides wide range and dynamical observation for the oceanic environments.In order to make it possible for the AUVs to take a prolonged task without surfacing up,the Docking Station is designed and plays an essential role.This paper focuses on the research on the navigation of AUV during its docking process.In this paper,a novel navigation method based on information fusion with RSS and monocular vision is proposed,and a low-cost navigation system is designed to achieve the full state estimation of AUV during its docking.As for the underwater localization based on RSS,the model of electromagnetic wave attenuation in water is firstly studied and the mathematical expression is deducted,which is verified by the HFSS simulation.The framework of Extended Kalman Filter is introduced and based on that,the localization algorithm with RSS is designed.Moreover,calibration and localization experiments are conducted in the lab’s pool,which favors the proposed localization algorithm.As for the bearings-only tracking algorithm with one camera and one light,the propagation characteristics of lights in the water and the underwater image formation model are studied.The automatic exposure algorithm and the Otsu image segmentation technology are proposed to get a clear underwater picture.Then,the procedure for the tacking algorithm is designed and a pool test was conducted to verify the theory.As for the navigation algorithm based on information fusion technology,the JDL model for the underwater docking is proposed,and specifically,the object assessment and sensor management parts are fully explained.The kinematic model for underwater docking is designed and the guiding algorithm is proposed with EKF framework based on information fusion.And also,a simulation experiment for AUV docking is conducted.Finally,this paper gives an overview of the underwater docking system,which focuses on the design of the subsystem of underwater communication.The software and hardware design for the navigation system are introduced and the docking experiment is conducted in the pool to test the information-fusion based navigation algorithm.