Artificial Intelligence to Predict the Risk of Ship Collision

Post sponsored by Cria de Caracoles Learn to Create a Nature Friendly Home Based Business


The system created by Fujitsu and tested in Singapore could detect incidents 15 minutes in advance

Fujitsu has announced the results of a field test conducted with the assistance of the Maritime and Port Authority of Singapore to analyze the risks of maritime traffic in the Singapore Strait . The Japanese multinational has demonstrated the utility of a new technology for predicting ship collision risk , which exploits the power of artificial intelligence (AI) to predict failures between ships.

With the AI Fujitsu Human Centric AI Zinrai technology ┬ádeveloped by Fujitsu Laboratories Ltd, this solution can detect ship collision risks and predict areas where risks are concentrated as ‘critical points of dynamic risk’ . This technology has the potential to be implemented in a Vessel Traffic Service system (VTS systems) to help maritime controllers proactively manage traffic , in order to improve the safety of navigation.

The Japanese multinational will offer in 2020 these services that incorporate this technology for maritime traffic controllers and ship operators.

To manage maritime traffic, the anti-collision warning is relied upon in VTS systems, which use the closest approach point techniques to detect and notify ships when they approach each other unusually . However, as the operations and interactions of ships become more complex, “the ability to detect and predict the movements of ships in advance, especially in areas of high density traffic such as Singapore, is key to manage and reduce the risks of collision. “

This dynamic risk hot spot detection technology could discover risks up to 15 minutes in advance , which would allow staff to take specific measures to avoid an incident.

Artificial Intelligence to Predict the Risk of Ship Collision

The tests of this system have been made in the Straits of Singapore. / Cameron Smith / Unsplash

System tests Based on the past traffic data for the Singapore Strait provided by the Port Authority, Fujitsu took advantage of its collision risk prediction technology to extract information, such as collisions or near misses, involving multiple vessels, as well as as examples of development of active zones of dynamic risk. The prediction technology was compared with human operators. Fujitsu’s risk detection technology demonstrated the possibility of quantifying risk in more detail, rather than detection by human operators. In addition, it was also able to detect scenarios in which the collision risk could be overlooked and warn of a possible incident.


Source: Innovadores La Razon