Conference moderator Professor Volker Bertram from World Maritime University didn’t beat around the bush to demystify the term ‘Artificial Intelligence’ (AI) in his opening remarks: “If we use a human brain for reference, AI – which doesn't know curiosity or eagerness to acquire knowledge – is basically an idiot, albeit an extremely useful one." At the Maritime Future Summit during SMM DIGITAL, two expert panels – each followed by a discussion round – exhaustively explored questions such as how the maritime industry may benefit from AI, and what challenges are associated with Deep Learning and Big Data. The cooperation partner of this event was once again the industry magazine Hansa.
Dr Pierre Sames, Group Technology and Research Director at DNV GL – Maritime, mentioned two practical AI application examples: Smart algorithms can be used to determine the service life and reliability of ship batteries; and the classification society uses big data to develop corrosion damage predictions. According to Sames, these new technologies enable ship operators to benefit from improved fleet performance and more efficient fleet management. However, smart applications are not self-propelling. "The trustworthiness of AI systems is not very different from that of a leader or an expert to whom we delegate our authority to make decisions,” Sames pointed out. This is why DNV GL – Maritime scrutinises AI systems used by its customers ("software testing software") and has developed a framework for secure application of data-driven algorithms.
Artificial intelligence is used in modern shipbuilding, as well. Examples were provided by Rodrigo Pérez Fernández from the technology company Sener: algorithms are used to optimise the design of module blocks or to determine the most efficient design for complex piping systems running throughout the entire ship.
Smart, clean and autonomous
Pierre Guillemin, Vice President Technology at Wartsila Voyage, highlighted how AI will revolutionise propulsion technology.The Finnish ship engine manufacturer will use digital technologies such as Smart Shipping to achieve significant ship efficiency improvements while adapting existing safety standards from the automobile industry. But Guillemin was also clear about the challenges and hurdles facing the digital transition in the shipping sector, including patchy network connectivity at high sea, the lack of consistent standards and regulations, and cyber crime hazards.
These are the reasons the industry has been rather slow to embrace the concept of autonomous or unmanned shipping, he indicated. Oskar Levander, Senior Vice President Business Concepts at Kongsberg Maritime, reported about current projects and milestones. Apart from highlighting the world's first remote-controlled (car) ferry "Falco", he presented the innovative "Automatic Crossing System" which enables ferries to commute between ports autonomously. AI technology makes these passages both efficient and economical. "A smart combination of radar, cameras, advanced sensor technology and AI algorithms avoids collisions," said Levander. This approach is commonly referred to as "situation awareness".
The same topic was addressed by Jilin Ma in his presentation "With the coming of the Maritime Revolution, is the unmanned bridge available?” Ma, Senior Engineer of electrical and electronic engineering at the Chinese classification society CCS, concluded: “While the technologies are steadily improving, bridge systems operating autonomously are not ready for practical use at this time."
In general, autonomous or remote-controlled systems are not suitable for every ship type. "Autonomous shipping technology will be used in feeder ships, ferries and other vessels operating in coastal areas. But for large containerships and tankers, this technology is not economically feasible," said Kongsberg expert Levander. What is more, autonomous propulsion systems must always be designed with full redundancy for safety reasons, he added. "This would be enormously cost-intensive in larger ship types, a fact that makes this technology doubly unattractive."
Prof. Annie Bekker from the South African Stellenbosch University reported about a true endurance test for artificial intelligence. The state-of-the-art research icebreaker "S.A Agulhas II", equipped with the latest in AI technology, including sensors and cameras for measuring ice thickness, was taken by its crew to one of the most remote and dangerous regions on earth. The goal of the mission was to test the reliability of the digital navigation assistance systems under extreme conditions. Her conclusion: "While AI technology is enormously helpful, you can never trust it blindly. You always need both, Artificial Intelligence and human creative thinking."