What is the 'Mayflower Autonomous Vessel', a fully autonomous operating vessel that combines IBM's AI technology, etc.?
The Mayflower Autonomous Ship , a fully autonomous vessel with no captain or crew, is scheduled to cross the Atlantic Ocean from September 2020 as part of a sea trial. The technology used for the Mayflower autonomous ship is expected to make significant changes in the shipping industry and to collect data on the ocean.
IBM News Room-Mayflower then and now
Sea Trials Begin for Mayflower Autonomous Ship's 'AI Captain'-Mar 5, 2020
You can see the operation image of what a Mayflower autonomous ship is like from the following movie.
Entering the Mind of the Mayflower-YouTube
A fully autonomous vessel developed by the ocean research institute Promare and IBM was named 'Mayflower Autonomous Vessel' to commemorate the 400th anniversary of the first Mayflower voyage in the United States in 1620.
Mayflower autonomous ship, United Kingdom of Plymouth from the United States of Plymouth up, is expected to cross the Atlantic at about the same route as the root of the Mayflower was followed in 1620. If the Mayflower Autonomous Vessel that can carry no humans succeeds, the Mayflower Autonomous Vessel will become the first large-scale fully autonomous vessel to cross the Atlantic Ocean, promoting commercial autonomous vessel development and transforming ocean research. It is expected to help.
As of 2020, there are many self-driving vessels that can be autopiloted. However, it is said that it is difficult to operate unmanned because many of the auto-operated vessels cannot respond flexibly and rely heavily on the judgment of the pilot. The Mayflower Autonomous Ship uses AI developed by IBM and an
The Mayflower Autonomous Ship development team spent about two years training the AI of the Mayflower Autonomous Ship using more than one million images taken in Plymouth Sound Bay, England. To meet the processing requirements of machine learning, IBM Power AC922 , a platform with IBM Power9 CPU and NVIDIA V100 Tensor Core GPU, has been adopted. As of March 2020, the AI onboard the Mayflower Autonomous Ship can automatically detect and identify obstacles such as ships, buoys, land, and breakwaters.
The Mayflower Autonomous Ship uses an edge computing system because it is difficult to connect to a broadband network that can transfer a large amount of information during a voyage. The Mayflower Autonomous Ship is equipped with multiple NVIDIA Jetson AGX Xaviers , which are single-board computers.By processing the data locally on NVIDIA Jetson at sea, the speed of AI's decision making is improved and the data onboard the ship is increased. We are reducing the amount of flow and storage.
“Edge computing is a very important technology that enables autonomous operation at sea. Mayflower autonomous vessels sense the environment, judge situations and make quick decisions, and act as quickly as possible. And yet, you have to do it while protecting your data from cyber-attacks, etc. IBM's edge computing solutions are designed to support
Mayflower Autonomous Ships not only operate according to the goal of 'from Plymouth in the United Kingdom to Plymouth in the United States in the shortest time', but also use IBM's decision automation system ' Operational Decision Manager (ODM) '. It is also designed to comply with the Convention on the Prevention of Collisions at Sea (COLREGs Convention) and the International Convention for the Safety of Life at Sea (SOLAS Convention) . It is also designed to receive weather forecast data from the Weather Company and perform operations in consideration of the weather, even for weather that has a major impact on voyages.
As of March 2020, the construction of the Mayflower autonomous ship itself is in the final stages in Poland.
Also, in the marine test conducted from March 2020 to May 2020, AI, which will be installed on the Mayflower autonomous ship, will be installed on a manned ship and a test of data input such as radar, GPS, navigation system will be conducted. I will. After April 2020, image processing, edge computing, and autonomy tests will be conducted.