A Decision Support Framework for Collision Avoidance of Unmanned Maritime Vehicles

Recently there has been considerable interest in the development of Unmanned Surface Vehicles (USVs) due to the increasing demand in a number of maritime applications. One of the main challenges for unmanned (and even manned) vessels is detecting and avoiding the (static and dynamic) obstacles that may appear during a mission. This paper presents a practical solution for obstacle detection and avoidance mainly for USVs. More importantly, an integrated decision support framework is proposed which provides risk assessment as an integrated feature to the path planner. The hardware platform consists of a high definition video camera and a laser range sensor mounted on a pan and tilt device to detect obstacles in the vicinity of the ship. The performance of the proposed system is extensively investigated in a number of virtual maritime environment scenarios. The results reveal that the proposed system is able to detect multiple dynamic obstacles. Additionally, the proposed system provides a real-time visual interface (similar to radar screen) for the captain which includes recommended actions to avoid the obstacles in accordance with marine rules of the road.