Applying cognitive situation awareness to collision avoidance for harbour last-mile area safety

Maneuvering ships in harbour areas is recognized as one of the most complex activity to be carried out Le (An automatic control system for ship harbour maneuvers using decoupling control. In: IEEE International Conference on Robotics and Automation, 2001. Proceedings 2001 ICRA 2:2108–2113, 2001) as a consequence of all information needs related to safety maintenance during guidance and control activities. By taking advantage of existing results, in this work we argue how situation awareness could be a leading and valuable technology to support decision making activities in this application. In particular, we propose the usage of a decentralized, autonomous, agent-based cognitive situation awareness framework, yet successfully experimented for airport safety monitoring and specifically modified to be applied to the maritime domain. The proposed framework exploits sensor network acquired streamed data management activities, together with semantic technologies and uncertainty management methods. The resulting architecture is aimed at monitoring ongoing situations and forecasting potential collisions, identifying involved entities and relations in order to improve the ability of maneuvering operators in assessing the current situations and their possible evolutions. Results and performances will be verified at the Port of Salerno, in order to show that information supporting inferred situations comply with real needs of decision making activities to prevent collisions during daily operations.