Font Size: a A A

Research On UUV Seafloor Terrain Tracking Method Based On Dynamic Path Generation

Posted on:2021-03-12Degree:MasterType:Thesis
Country:ChinaCandidate:Q Q PiFull Text:PDF
GTID:2492306047997509Subject:Control Engineering
Abstract/Summary:PDF Full Text Request
Unmanned Underwater Vehicle(UUV)is an underwater vehicle that is unmanned and independently accomplishes specific mission tasks by relying on self-loading control system.As humans continue to explore the deep sea,unmanned underwater vehicles that adapt to the deep sea environment have emerged.Underwater unmanned vehicles are mainly used to complete deep sea environmental intelligence collection,topographic surveys,shipwreck salvage,underwater search and rescue task.In particular,when carrying out the seafloor topography survey task,the UUV needs to keep a constant altitude with the seafloor terrain for tracking and navigation to ensure the reliability of the detected terrain information.At this time,the operating mode is to track and navigate the seafloor terrain.Due to the complex and rugged topography of the seafloor,it is difficult for the UUV environmental sensor to detect the seafloor changing terrain accurately,and the detection range of the sensor is limited.Therefore,it is of great theoretical significance and practical value to study the dynamic planning and generation of the UUV seafloor terrain tracking path of unknown seafloor terrain.This paper studies the dynamic planning of the navigation path and the online generation of the path during the UUV seafloor terrain tracking process,and fully considers the problem of the mismatch between the execution cycle of the planning module and the execution module,the discrete problem of the sonar detection points,and the real-time update of the terrain information.A three-layer closed-loop receding horizon planning and tracking framework path dynamic planning method and a path generation method based on detection information fitting are proposed.First of all,in order to clarify the task requirements in the path dynamic planning generation and stability control,the kinematics and dynamics modeling analysis of the UUV vertical plane was completed.Considering the UUV’s detection requirements for terrain information,a multi-beam forward-looking sonar model was established,and UUV’s detection modeling of external terrain information was completed.And analyze the actual seafloor terrain characteristics,and establish four typical seafloor terrain models.Secondly,it is aimed at the unified coordination problem of shorter controller execution cycle and longer path dynamic planning generation cycle.A three-layer closed-loop feedback structure for planning,tracking,and control is proposed.The feedback loop is designed according to different functions to separate the tracking,control,and planning modules of different frequencies.Based on the receding horizon planning and tracking framework,a dynamic planning algorithm is designed,and a reasonable planning range and planning timeare selected to ensure the realization of the receding horizon path planning generation method.Finally,in the process of generating planned paths,based on the sonar detection information,the advantages and disadvantages of Bessel fitting,polynomial fitting and Gaussian fitting are compared,and the method with the best fitting effect is selected Complete the online generation of the sailing path.The two-dimensional seafloor terrain was established through a piecewise function,and the effects of different speeds,different forward-looking sonar opening angle configurations and different path planning times on the seafloor terrain tracking effect were studied.Simulation experiment results show that the dynamic path generation method proposed in this paper can guide UUV to complete the tracking task of unknown seabed terrain,and the designed tracking control framework and path generation method have good practical value.
Keywords/Search Tags:Unmanned Underwater Vehicle, Near bottom terrain tracking, Receding horizon path planning generation, Detection information fitting
PDF Full Text Request
Related items