Target Capture And Motion Tracking Of AUV Targeting System

Ocean account for about 70% of the earth’s total area and are rich in resources.In order to maximize the use of marine resources,a series of underwater industrial robots have emerged.The underwater robot equipped with the autonomous underwater submarine(AUV)target capture and tracking system is of great significance to the exploitation and protection of the ocean.AUV target systems can be used not only in the military field(e.g.territorial defense,missile tracking,etc.),but also in civilian areas(e.g.submarine pipeline laying,marine environmental survey,etc.).Therefore,a new underwater autonomous submarine(AUV)target capture tracking system based on four-pulp propulsion is designed.First,on the basis of a clear understanding of the mechanics and kinematics principles of AUV,the corresponding AUV target capture tracking system was set up in the laboratory to simulate the actual operation of AUV sailing in the sea.The AUV target capture tracking system includes front-end image recognition system and back-end motion tracking system.The front-end image recognition system mainly includes USB camera and Raspberry Pi development board,and the detection algorithm used to capture the target image is Hof circle detection and YOLOv3-tiny algorithm.The construction of the front-end image recognition system model mainly includes the establishment of the reference coordinate system,the establishment of the camera model and the dynamic model of the AUV,so as to derive the overall transfer function of the system.Secondly,The back-end motion tracking system mainly includes four-rotor motors,balance bars and inertial navigation.Aiming at the low control accuracy and slow response speed of the AUV target tracking system,this paper intends to use a dual closed-loop PID to control the attitude movement of the AUV.The difference between the center coordinates of the target and the center coordinates of the image identified by the front-end image recognition system is the input of the inner ring of the dual closed-loop PID controller,and the output of the ring PID is the input attitude angle of the inner ring.The outer ring continuously calculates the output angle of the coordinate difference between the two,while the inner ring dynamically adjusts the angle to stabilize the output,ensuring that the center point of the target coincides with the center point of the image in order to achieve AUV target capture tracking.Because the inner ring calculation requirement must be fast and accurate,consider the advantages and disadvantages of various algorithms In this paper,the inner ring adopts the variable domain fuzzy PID control algorithm.Finally,on the basis of the above theory and model,this paper has carried out simulation simulation and experimental verification.(1)Simulation simulation: Through MATLAB simulation simulation,the simulation results show that the variable domain fuzzy PID algorithm has great advantages in terms of speed and accuracy compared with other control algorithms.(2)Experimental verification: in the laboratory for the AUV target capture tracking system set up in this paper for hardware and software experimental verification,capture tracking results map shows that the AUV target targeting system for static and moving targets have a good capture and tracking effect.