Ship collision is the main type of maritime accidents, which causes great losses on human lives and economy, and brings negative impacts to the maritime environment. In crowded waters such as the sea area near a seaport, multiple ships encountering situations happen frequently. While several methods have been proposed for solving multiple ships collision avoidance problem, most research focuses on the safety guarantee and time-availability of anti-collision decisions, and less attention is paid on improving the efficiency of collision avoidance maneuvers. This paper proposes a rolling horizon optimization approach for multiple ships from a global optimal perspective, with the aim to minimize the time costs and course angle alterations of the anti-collision operations. A ship maneuverability model is used to make predictions and calculations of inter-ship collision risks, upon which an overall optimization problem is formulated. Continuous time is divided into a set of discrete time slots, and an iterative optimization procedure is carried out to determine the optimal course angles for the ships at each time slot. A PID heading controller is designed to implement the optimal course angles. When collision risks among ships no longer exist, the optimization procedure terminates. To evaluate the performance of the proposed approach, simulation experiments regarding 7-ships and 12-ships encounter situations are carried out.@en

Human experience is regarded as an indispensable part of artificial intelligence in the process of controlling or decision making for autonomous cargo ships. In this paper, a novel Deep Q-Network-based (DQN) approach is proposed, which performs satisfactorily in controlling a cargo ship automatically without any human experience. At the very beginning, we use the model of KRISO Very Large Crude Carrier (KVLCC2) to describe a cargo ship. To manipulate this ship has to conquer great inertia and relatively insufficient driving force. Subsequently, customary waterways, regulations, conventions are described with Artificial Potential Field and value-functions in DQN. Based on this, the artificial intelligence of planning and controlling a cargo ship can be obtained by undertaking sufficient training, which can control the ship directly, while avoiding collisions, keeping its position in the middle of the route as much as possible. In simulation experiments, it is demonstrated that such an approach performs better than manual works and other traditional methods in most conditions, which makes the proposed method a promising solution in improving the autonomy level of cargo ships.@en

Human experience is regarded as an indispensable part of artificial intelligence in the process of controlling or decision making for autonomous cargo ships. In this paper, a novel Deep Q-Network-based (DQN) approach is proposed, which performs satisfactorily in controlling a cargo ship automatically without any human experience. At the very beginning, we use the model of KRISO Very Large Crude Carrier (KVLCC2) to describe a cargo ship. To manipulate this ship has to conquer great inertia and relatively insufficient driving force. Subsequently, customary waterways, regulations, conventions are described with Artificial Potential Field and value-functions in DQN. Based on this, the artificial intelligence of planning and controlling a cargo ship can be obtained by undertaking sufficient training, which can control the ship directly, while avoiding collisions, keeping its position in the middle of the route as much as possible. In simulation experiments, it is demonstrated that such an approach performs better than manual works and other traditional methods in most conditions, which makes the proposed method a promising solution in improving the autonomy level of cargo ships.@en