1960s the US Defense Advanced Research Projects Agency creates the first AI systems
1990s Subfield concepts fragment AI into machine learning, neural networks, computer vision, and natural language processing
2010s Corporations such as Facebook, Google, and Amazon pioneer AI business applications, using it to power search results, improve logistics, and more.
The prospect of a ship captain being replaced by artificial intelligence is closer than you might think. In some cases, such as small vessels along the Scandinavian coast, it is a reality. In the short term, the technology is not expected to replace human crews. Rather, most experts expect AI will complement traditional human roles throughout the industry. With machine learning taking on complex tasks to increase productivity and reduce failures, the benefits to the entire marine industry are apparent. We’ve already covered innovative tech including autonomous vehicles, AI assisted undersea mapping, and condition monitoring. But on the horizon, we can see AI being applied in other ways across the marine industry, from dry dock, to harbour, to sea.
When today’s cutting-edge vessels leave dry dock for a lifetime of service, hundreds of sensors come along for the journey. These instruments, some of which are smaller than a pen, constantly check on the status of many of the ship’s moving parts, gathering data in real-time. But can a human make sense of this data in order to make important decisions about predictive maintenance? Not a chance. Machine learning and deep learning algorithms have stepped forward to carry this load.
In dry dock, machine learning is playing an important role. It automates radiography testing, which uses x-rays or gamma-rays to detect imperfections. The tests can also improve the quality of welds, and the speed that AI brings to the process can increase build-times.
Artificial intelligence promises to revolutionize predictive maintenance. In this SKFstronger webinar, SKF AI Offering Manager and author Eitan Vesely explains how insights provided by machine learning fit into Maintenance 4.0.
The IOC is committed to reducing the marine industry’s carbon footprint, and artificial intelligence has a role to play there, too.
Just as the sextant gave way to GPS, Stena Line targets 2021 for these AI systems to begin assisting vehicle navigation. Its computer assisted course calculations should improve fuel efficiency on voyages by helping to find travel windows with good weather and low traffic. And as shipping emissions regulations take effect, these models will also help minimize environmental impact.
Increased safety, another hot topic, has been an industry goal since goods were first shipped by sea, and AI first mates may soon help vessel operators to recognize ships, bridges, and other hazards. Recognizing this, the German government is funding research that uses sensors and autonomous technology to guide ships in Hamburg. Human error is responsible for 75% to 96% of marine accidents, and AI-enabled navigation has the potential to dramatically reduce that number, which is good news for seafarers as well as owners and operators.
Harbours are being taken over by artificial intelligence, too. A little over half of terminals across the globe are trialing AI use in container handling equipment assignments, decking systems, or stowage vessels. Companies offering solutions that help handling cargo across the globe are on the front lines, and just as the larger logistics networks have benefited from machine learning and automation, so will local ports.
Set course for the future
From dry dock, to harbour, to blue water, artificial intelligence is already changing the marine industry. But what lies over the horizon?
Plans for autonomous vessels are already underway, with the ultimate goal of being fully autonomous ships. The first step toward that goal began with a historic twist: an AI Captain prototype at the helm of a ship named Mayflower. The trials start later this year, with the goal of crossing the Atlantic Ocean. This IBM and Promare project will feature technology that helps the vessel avoid unexpected hazards, among other things.
The tech is helping law enforcement, too. By combining information from radar and satellite data, machine learning algorithms can identify illegal activity at sea with so-called dark vessel detection.
This nearly 60-year-old AI technology is bringing long sought-after breakthroughs and unexpected changes with it. For the marine industry, this is only the beginning.