Resarch On UUV Visual Guided Control For Underwater Docking

With the world’s population growing at an explosive rate,the consumption of Earth’s resources is accelerating.This has prompted more and more people to focus on food security and diversification,seeking ways to develop and produce more resources within their own territories.As a maritime power,researchers are increasingly focused on how to further develop and utilize marine resources.In the field of ocean engineering,the research of underwater unmanned equipment,represented by unmanned underwater vehicles(UUVs),has been a main area of interest for several decades.UUVs are widely used for submarine infrastructure inspection,environmental monitoring,and marine surveying and mapping.However,during underwater operations,the energy carried by UUVs is very limited and cannot be supplemented by self-energy.Therefore,in the face of large-scale and complex underwater aquaculture industries,where there are high control requirements,large working ranges,and long-term operations,UUVs are powerless.Rapid docking and recovery of UUVs,energy supplementation,and data transmission are important technical means to expand the application of UUVs in large water aquaculture industries.Based on the above background,this paper focuses on the development of autonomous underwater docking for a fully actuated underwater robot(the Blue ROV2).The research aims to enable Unmanned Underwater Vehicles(UUVs)to perform autonomous docking underwater.Through thorough investigation,a vision-based autonomous docking recovery scheme is proposed and tested in combination with specific scenarios to ensure its stability and reliability.The results provide essential technical support for the deployment of multiple recycling stations for underwater docking and recycling,facilitating comprehensive monitoring of the aquaculture process in the large water aquaculture industry.The specific research objectives are as follows:Firstly,the paper begins by discussing the core task of autonomous underwater docking,focusing on conducting relevant technical research and analyzing the challenges faced by current research in underwater visual guidance.To facilitate subsequent research and control program development,the paper introduces the hardware and software systems of the Blue ROV2 underwater robot,which is the research object.Furthermore,the paper presents dynamic modeling,kinematic modeling,and mathematical modeling of the propeller system by referring to the underwater robot theoretical model proposed by Norwegian scholar Fossen and the propeller system of UUVs.Through the analysis of relevant technologies,a solution was designed for the overall visual guidance docking process,laying the theoretical foundation for subsequent work.Secondly,the paper addresses the challenges presented by the complex underwater scene and characteristics of underwater visual guidance docking.To overcome these challenges,a large number of underwater docking target data were collected using the UUV’s own camera,and a real-time detection method based on YOLOV5 s was designed.To address the distortion problem of the monocular camera equipped with the robot during the docking process,the paper calibrates the camera’s distortion in water and internal and external parameters using Matlab software.With the successful detection of the docking target and the acquisition of the camera’s internal and external parameters,the EPn P algorithm was then utilized to calculate the real-time attitude between the UUV and the target docking device.This calculation provides the necessary prerequisite for the realization of visual guidance docking control.Then,to address the tracking control problem in the docking process,the paper compares the advantages and disadvantages of existing control technology.Based on the solved attitude information,the paper designs a terminal sliding mode trajectory tracking control algorithm using LOS guidance law,sliding mode control algorithm,RBF neural network universal approximation theory,and Lyapunov stability theory.Furthermore,the paper designs different tracking scenarios and disturbance types using Matlab software to carry out horizontal tracking control and three-dimensional space tracking control.The simulation results demonstrate that the designed tracking controller has a stable and accurate tracking effect and strong robustness against interference.Finally,the paper implements the designed visual guidance scheme on Blue ROV2 through secondary programming development,integrating the designed target detection,attitude calculation,and tracking control algorithm into the shore-based control program.The paper conducts relevant tests in the ship pool and further improves and optimizes the program according to the actual control situation.After multiple tests and improvements,the underwater robot can complete underwater docking and recovery in a relatively stable and accurate motion state under visual guidance.This paper verifies the practical operability of the designed visual guidance docking scheme and provides ideas and methods for the research and design of the docking scheme in the large water aquaculture process.Additionally,this paper provides a technical reference for future underwater unmanned navigators to complete underwater operation tasks more safely,reliably,and efficiently on large voyages.