A ship for the AI captain

Autonomous shipping and ocean research

Jonathan Batty and Rosie Lickorish of IBM Europe describe the company’s role in the Mayflower Autonomous Ship research project, and the challenge of the vessel’s transatlantic voyage, scheduled for spring 2021

With human activity damaging ocean health, disrupting climate patterns and destroying ecosystems, our oceans are warmer, more polluted, more acidic, more depleted of life, and more unpredictable than ever before. While the global importance of the Amazon rainforest is undoubted, the ocean, and the oxygen-emitting phytoplankton that live in it, can make a greater case for being called the ‘lungs of the planet’, producing about half of the air we breathe. Marine and coastal industries worth approximately $3 trillion a year depend directly on the ocean. For these reasons, the United Nations has proclaimed the 2020s to be the Decade of Ocean Science for Sustainable Development. Bringing together a range of scientific disciplines, policy makers and industry partners, the initiative will provide a framework for better understanding and managing marine environments. The need for oceanographic data and the ships that collect them has never been greater. But ocean exploration is labour intensive, expensive and sometimes dangerous.

The Mayflower Autonomous Ship

CGI visualisation of the MAS at sea
CGI visualisation of the MAS at sea. The vessel’s energy source is solar power with diesel generator backup. Photo: IBM

The growing maturity of autonomous technologies offers a path forward. IBM, answering the UN’s call for collaborative partnerships between the private and nonprofit sectors, is supporting marine research organisation Promare by providing the advanced technologies for the Mayflower Autonomous Ship (MAS). Four hundred years after the original Mayflower sailed from Plymouth for New England, MAS is using machine learning and energy from the sun to follow a similar course, autonomously, gathering and transmitting vital data to expand our understanding of the factors influencing ocean health.

Shuttleworth Design’s initial 32m hull concept has now evolved into a state-of-the-art 15m trimaran, with a maximum speed of 10 knots. The hull, built by Aluship shipbuilders at Gdańsk, is of aluminium and composite materials; it weighs 5 tonnes (3 per cent of the weight of the original Mayflower). It was shipped to MSubs in Plymouth for outfitting. The choice of a trimaran hull form was based on efficiency, resilience and stability in rough seas. This cutting-edge vessel, which took to the water for the first time on 16 September 2020 and will be fully operational by the end of October, represents not just the future of marine exploration but a fundamental shift in our overall approach to oceanographic research. The transatlantic crossing is planned for 19 April 2021. Ahead of that MAS will undertake a number of smaller missions and research projects.

Using an innovative new ‘AI Captain’, built by Promare’s engineers and combining IBM’s AI, ‘edge technologies’ (locating computation and data storage near where it is needed) with data from the Weather Company, MAS can identify and avoid hazards as it self-navigates at sea. IBM and Promare researchers are also pioneering new AI-powered approaches to the collection and on-site analysis of samples and ocean data, for better understanding of everything from water quality to what whale songs can tell us about marine mammal populations. One device includes a special version of an electronic, AI-assisted “tongue”, with an array of sensors facilitating rapid analysis of ocean water.

In the future, squadrons of autonomous research boats, drones and submersibles could spend as long as six months at sea, constantly collecting data, looking beyond seasonal variability to better identify and respond to long-term trends. They will work in tandem with human oceanographers, but allow these scientists to spend more time on data interpretation and consequent action, rather than data collection.

Beyond the research itself, MAS will help fulfil another important objective: engaging the public. Ultimately, it will take a collective effort to shift the current environmental trends. Through the story of the ship, the fusing of past, present and future, and the information shared globally about its mission, MAS will help to increase public awareness of issues related to ocean health. A new mission for the ship, MAS400.com has been developed to deliver on that objective.

It’s not too late

Ocean science has made huge progress in the last 150 years, but there are some significant gaps. Today less than 20 per cent of the ocean floor is mapped, much of it at very low resolution. The ocean has a powerful ability to regenerate itself, but unless we improve our understanding of ocean health and use developing technology to help us do so sustainably, we risk setting off a chain of degradation from which it may be impossible to recover. The UN Seabed 2030 Project recently awarded to the Nippon Foundation – General Bathymetric Chart of the Ocean (GEBCO) consortium will complete the entire mapping of the ocean floor by 2030.

In spite of the damage, it is not too late to save our seas. For the first time in history, we have at our disposal a set of technologies that can completely transform our ability to – in the UN’s words – ‘gather the data we need, for the ocean we want’.

A ship making its own decisions

Graphic representation of the MAS “AI Captain”
Graphic representation of the MAS “AI Captain”, detecting, analysing and responding to a hazard in its path – the results of a collision between a container ship and a fishing vessel, with debris in the water. Photo: IBM

In facing some of the most challenging circumstances on the planet, with no human captain or crew on board, MAS will rely on IBM’s advanced AI and edge computing systems to sense, think and make decisions at sea. Over the past two years, while MAS’s hull has been under construction, the Mayflower team has been training the ship’s AI models, using over a million nautical images collected from cameras in Plymouth Sound, as well as open source databases. Now, using IBM’s computer vision technology, the Mayflower’s AI Captain should be able, independently, to detect and classify ships, buoys and other hazards such as land, breakwaters and debris. The Mayflower will not have access to high-bandwidth connectivity throughout its transatlantic voyage. It will use a fully autonomous IBM edge computing system to process data locally, increasing the speed of decision making and reducing the amount of data flow and storage on the ship.

‘Edge computing is critical to making an autonomous ship like the Mayflower possible,’ says Rob High, VP and CTO for Edge Computing, IBM. ‘The Mayflower needs to sense its environment, make smart decisions about its situation and then act on these insights in the minimum amount of time – even in the presence of intermittent connectivity, and all while keeping data secure from cyber threats.’

The AI Captain will also draw on IBM’s rule management system to follow the International Regulations for Preventing Collisions at Sea (COLREGs), as well as recommendations from the International Convention for the Safety of Life at Sea. As weather is one of the most significant factors affecting the success of the voyage, the AI Captain will use forecast data from The Weather Company to help make navigation decisions. A Safety Manager function will review all of the AI Captain’s decisions to ensure they are safe – for the Mayflower, and for other vessels in its vicinity.

How the Mayflower senses, thinks and acts at sea

Let’s assume that the Mayflower is in the open ocean, approaching Cape Cod, with no current satellite connectivity. Ahead is a cargo ship which has had a collision with a fishing vessel and spilt some of its load. In this hypothetical scenario, the Mayflower’s AI Captain will use the following technologies and processes to independently assess the situation, and decide what action to take:

SENSES (assesses current environment, identifies hazards):

  • Radar detects multiple hazards in MAS’s path, 2.5 nautical miles ahead.
  • Onboard cameras provide visual input to IBM computer vision system, which identifies hazards as: a cargo ship, a fishing vessel and three partially submerged shipping containers floating in the water.
  • Automatic Identification System (AIS) provides specific information about the cargo ship’s class, weight, speed, cargo, etc.
  • GPS Navigation System provides MAS’s current location, heading, speed and course.
  • MAS’s nautical chart server provides geospatial information about its chosen route.
  • Weather data is provided by The Weather Company.
  • Attitude Sensors assess local sea state (how MAS pitches and rolls due to waves).
  • Fathometer provides water depth measurements.
  • Vehicle Management System provides operational data such as MAS’s battery charge level, power consumption, communications, science payloads etc.

THINKS (evaluates options)

  • IBM Operational Decision Manager (ODM) evaluates COLREGs with respect to the other vessels in the vicinity and generates a risk map indicating an “unsafe” situation ahead.
  • MAS’s AI Captain takes in the ODM recommendation, computer vision input, current and forecasted weather, and assesses several options to avoid hazard.

ACTS (chooses best actions and instructs vessel)

  • AI Captain determines that the best action for MAS is to steer to starboard to avoid the unexpected navigation hazard.
  • MAS’s Safety Manager verifies the decision as safe.
  • AI Captain instructs MAS’s Vehicle Management system to change course and speed.

As the ocean is an ever-changing dynamic environment, the AI Captain will constantly re-evaluate the situation and update the Mayflower’s course.

Besides helping to transform the future of marine research, the Mayflower’s mission will further the development of commercial autonomous ships in a market that is set to grow from $90 bn today to over $130 bn by 2030.

At IBM Europe, Jonathan Batty is Head of Content and Storytelling, and Rosie Lickorish is a software engineer.