Mitsui O.S.K. Lines (MOL) has teamed up with Japan’s National Maritime Research Institute, the Tokyo University of Marine Science and Technology, and YDK Technologies to conduct a joint study on collision avoidance technologies for the shipping sector.
The goal of the study is to develop an advanced navigation support system that will support autonomous collision avoidance, using rule-based artificial intelligence and deep reinforcement machine learning algorithms.
MOL says that these technologies will enable onboard systems to estimate several Obstacle Zones by Target (OZT) among ships or objects at sea and propose a route that minimises the risk of a collision with those obstacles.
Demonstration testing with Tokyo University of Marine Science and Technology’s own ship, the Shioji Maru, has already begun, taking place in congested sea areas such as Tokyo Bay as a part of ongoing efforts to develop the group’s computational algorithms.
The tests to date have confirmed the system’s ability to estimate OZT targeting of ships in actual operation, MOL says, and its capabilities to develop and suggest avoidance routes in real time on board.
The partners will develop the system further with the aim of supporting medium-to-long-term strategies for avoidance navigation well before target ships pose a risk in congested sea lanes, taking into consideration the experience of officers and other personnel.