Splash247: AI watchkeeper put through paces in Mediterranean trial

British class society Lloyd’s Register has taken a closer look at AI-assisted navigation, completing a live vessel trial of Orca AI’s computer vision platform on a feeder containership operating in busy Mediterranean waters.

The five-day assessment focused on how AI can support bridge teams in real conditions, particularly when visibility is poor or traffic is dense—two areas where human workload and risk tend to spike.

The trial ran from Gioia Tauro in Italy to Marsaxlokk in Malta, covering 828 nautical miles and including high-traffic areas such as the Strait of Messina and port approaches. The system’s performance was measured against radar, AIS and visual watchkeeping.

During the voyage, the AI platform picked up close-range and low-signature targets that were not always visible on conventional systems, including small craft and non-AIS vessels.

It gives watchkeepers an extra layer of awareness in situations where detection gaps can lead to close encounters.

The Med remains one of the most demanding trading environments for bridge teams

Lloyd’s Register ship performance specialist Han Beng Koe joined the vessel as onboard assessor, tracking how the system performed in day-to-day operations.

“As the onboard assessor, I observed the demonstrated capabilities of AI-based computer vision within the operational environment. This provides a clear indication of the performance potential and scalable application of emerging technologies in maritime navigation systems,” he said.

The system, using Orca AI’s SeaPod units mounted on the bridge, combines day and thermal cameras to provide 360-degree coverage, acting as a digital watchkeeper that detects and tracks targets in real time.

Across the trial on a feeder containership, the platform logged 739 targets, many of them small or hard-to-detect objects. It achieved a 94% precision rate and 98.6% recall, while maintaining zero downtime throughout the voyage.

Lloyd’s Register said the project was not just about raw detection performance, but also about how the technology fits into real bridge operations.

“This significant project serves as an important reference point for data-driven system evaluations. It reflects our shared commitment to the adoption of novel technologies, at a time when decarbonisation and autonomy are becoming increasingly intertwined,” said Dipali Kuchekar, product manager for marine and offshore at LR.

The trial also placed a strong emphasis on human factors, with structured feedback collected from crew and workshops held to refine usability.

“From a human factors perspective, it is not just about what the technology can do. It is about how effectively it supports the human operator. These workshops demonstrated how structured feedback and user-centred design can play a critical role in shaping safer and more usable AI-enabled navigation systems,” said Stephanie McLay, team lead for human factors at LR.

For Orca AI, the results underline how quickly AI tools are moving from trials into everyday operations.

“What this trial shows is that AI-assisted navigation is no longer a future concept, it is already delivering measurable value in live operations. More than 1,200 vessels using Orca AI are evidence that earlier and more accurate detection, lead to more-informed decisions on the bridge, which lead to safer navigation. Trials like this pave the way for broader AI adoption in our industry on the journey towards autonomous shipping,” said Dor Raviv, chief technology officer and co-founder of Orca AI.

The Mediterranean remains one of the most demanding trading environments for bridge teams, with dense traffic flows, short sea routes, fishing activity and frequent port calls compressing decision-making time. Narrow passages such as the Strait of Messina and busy approaches across Southern Europe and North Africa leave little margin for error, making it a natural testing ground for technologies aimed at improving situational awareness and reducing navigational risk.