The marine research organization ProMare and IBM have developed an “AI captain” for the Mayflower Autonomous Ship (MAS), and by the end of this year, the plan is to cross the Atlantic without a crew.
The idea is that the boat will take roughly the same route as the original Mayflower did 400 years ago when it transported English emigrants, called ‘pilgrims’ to the United States. It’s not the first time a ship is claiming to be self-driving across the Atlantic, but Don Scott, CTO of the Mayflower Autonomous Ship, says their AI captain should be able to handle much more advanced situations than other similar autonomous boats have had.
“Using an integrated set of IBM’s AI, cloud, and edge technologies, we are aiming to give the Mayflower the ability to operate independently in some of the most challenging circumstances on the planet,” said Don Scott, CTO of the Mayflower Autonomous Ship.
The AI is currently being tested in the waters of the British Plymouth coast. If all goes well, it is planned to have the new version of Mayflower sealed from British Plymouth to US Plymouth in the state of Massachusetts in connection with celebrating Mayflower’s 400th anniversary in September.
It took the original ship more than two months to make this journey; now, its successor is expected to complete it in just 12 days. As is clear, much of it is because it will be much smaller and almost ten times faster (the Mayflower having a speed of 2.5 knots). Built-in carbon fiber and aluminum, this 15-meter long and five-ton ship is able to reach a maximum speed of 20 knots (37 km/h). To achieve this, it will implement a propulsion system that combines wind and solar energy, in addition to a diesel generator that will only be used if necessary.
In addition to this, the vessel will have four different sensors incorporated to measure the different variables of the ocean during its route. This will allow the collection of vital data for areas such as meteorology, oceanography, climatology, biology, the development of more autonomous navigation equipment, and also for studies related to marine pollution and environmental conservation.
For the past two years, the Mayflower team has been training the ship’s AI models using more than one million nautical images collected by cameras in Plymouth Sound Bay, as well as open-source databases. The team has also used an IBM Power AC922 and IBM Power AI Vision so that the AI Captain should be able to independently detect and independently classify ships, buoys, and other obstacles such as land, breakwater, and debris.