Research On The Control System And Sonar Target Tracking Of Autonomous Underwater Vehicle

At present,due to Autonomous Underwater Vehicle(AUV)'s advantages such as high flexibility,adaptability,safety,and high efficiency,it is widely used in such fields as underwater resource detection,underwater intelligence collection,underwater rescue and salvage,etc.This article combines with the 2015 international cooperation project of Zhenjiang,which need to develop an autonomous underwater robot for underwater exploration.Target recognition and target tracking research was conducted based on images acquired in AUV's sonar experiments,The specific research content is as follows:Firstly,based on the research status of autonomous underwater vehicles of domestic and foreign and the whole requirement of the project,proposed the whole development plan for the AUV's control system,we finish the ontology body design and selection of key equipment components of AUV.Then we carry out the design of AUV's hardware.The water surface control system includes a water control box and a laptop computer.The underwater control system is a system in which the PC104 industrial control board acts as the main control unit of the autopilot and the PC104 camera board processes camera images separately.At the same time,a control board is designed and manufactured,which includes sensor switch control circuit,communication relay circuit,leak detection circuit,etc.Secondly,AUV's software designed was conducted.The water surface control system monitoring software is based on MOOS platform and developed in esclip exploitation environment.On underwater control system's software,we have completed the design of the relevant program of the control board and reached the motion control functions such as dynamic positioning,depth orientation,and trajectory tracking.Subsequently,the sonar target tracking research was conducted based on AUV.The entire tracking process including sonar image enhancement,sonar target detection,ellipse feature extraction and recognition,target tracking.A three-frame difference method was used to detect the sonar target.A fast ellipse feature extraction algorithm based on Hough transform was proposed.The target tracking algorithm is studied and an improved particle swarm optimization particle filter algorithm is designed.The superiority of the algorithm to the conventional particle filter algorithm is verified by a simulation model.Finally,the prototype of the autonomous underwater robot was tested in lake experiments.The experiments including dynamic positioning,depth orientation,etc.The feasibility and effectiveness of AUV's control system were verified through experiments,the design ofAUV has reached the desired requirement.In addition,the sonar target recognition experiment is carried out.The typical sonar images are selected for target recognition,the validity of the target recognition algorithm proposed in this paper is verified.