Fukae Maru, a training vessel owned by Kobe University in Japan, was recently used to test a new AI-based collision avoidance and navigation software.
The software, developed by Nippon Yusen Kabushiki Kaisha (NYK) subsidiaries MTI and Japan Marine Science, determines safe and economical course selection in various circumstances by applying deep reinforcement learning, a machine learning method in which an autonomous program performs a vast number of voyage simulations to learn optimal collision avoidance strategy.
The system uses radar and AIS sensor information to recognise collision risks in ship navigation and automatically selects the optimal course. The selected course is then set as the target course of a vessel's autopilot for manoeuvring.
The tests were carried out in Osaka Bay.
