Control Design And Implementation Of Autonomous Underwater Vehicle

Sea cucumber,sea urchin and other marine organisms contain high nutritional value,which makes people’s demand for marine organisms increasing day by day.At present,marine organisms are mainly caught by hand,which is dangerous and inefficient due to the complex underwater environment.This paper addresses this issue by using underwater robot control technology combined with computer vision to complete the capture of relevant marine organisms.The main research contents are as follows:(1)According to the designed structure,parameters and its own motion characteristics,the fixed coordinate system and motion coordinate system are respectively established and coordinate transformation and parameter definition are carried out.The linear and rotational motion equations of underwater robots are deduced by rigid body motion theory.The static forces and thrust of the thrusters are analysed and the hydrodynamic forces are investigated using potential flow theory to provide a theoretical basis for motion control.(2)Put forward the overall design scheme of the control system.STM32L4 XX is selected as the main controller and the control circuit is designed.Select the main sensor and electronic components and design the circuit,and finally draw the PCB control board.PID control algorithm is used to control the attitude of underwater robot.The overall design of the software system mainly includes MCU software design,lower computer software design and upper computer monitoring interface design.(3)Multi-target detection of relevant Marine organisms through computer vision technology.YOLOv5 target detection algorithm and Py Torch deep learning framework are used for model training.The target Marine biological data set is annotated and processed by Mosaic,adaptive anchor frame and other techniques.The training results were evaluated by confidence degree,P-R curve and other evaluation indexes.(4)Make the underwater robot prototype,verify the rationality and feasibility of the control system design through the underwater robot attitude motion control test,image acquisition test and grasping test of the target marine life.