Research on Unmanned Surface Vessel Perception Algorithm Based on Multi-Sensor Fusion

At present, the data fusion algorithm for multi-sensor is not effective. This paper studies the perception technology of the Unmanned Surface Vessel (USV), mainly from three aspects: single-sensor object detection algorithm, object detection algorithm based on multi-sensor fusion, and multi-object tracking algorithm. A set of 3D object detection and tracking algorithm based on the image and original point cloud data is designed, which can obtain the situation information of the water surface target in real-time. In addition, the hardware and software design of the ROS-based USV system is implemented. It helps the boats to better realize the functions of positioning, navigation, and obstacle avoidance, and to meet the needs and complete the tasks under various environmental conditions. It has high engineering value and practical significance in realizing the autonomous driving function of boats.