Real-time obstacle detection for Unmanned Surface Vehicle

Abstract

This paper presents a real-time obstacle detection system for obstacle detection on the sea surface for Unmanned Surface Vehicle (USV). The system employs both monocular and stereo vision-based methods and offers the capacity of detecting and locating multiple obstacles in the range from 30 to 100 meters for high speed USV. Field tests have been taken and shown that the real-time obstacle detection system for USV can provide stable and satisfactory performance.