Design of Autonomous Obstacle Avoidance Unmanned Boat System for Wetland Monitoring

Abstract In view of the drastic reduction of wetland area, serious pollution, low efficiency of prior art and many factors, this paper proposes an autonomous obstacle avoidance unmanned boat system design for wetland monitoring. The system adopts D* algorithm and improved LOS guidance law to realize unmanned ship track control and global path planning, and combines Yolov3 algorithm for target recognition and partial data acquisition during navigation to achieve autonomous obstacle avoidance navigation. The sensor is used to collect the wetland environment information and upload it to the shore monitoring centre to complete the automatic data collection and monitoring of the wetland. The system can improve the efficiency and quality of wetland monitoring, reduce labor costs, and can be widely used in wetland monitoring automation, marine environment monitoring, and cloud service platforms.