Offshore safety enhanced by AI detection system from Zelim

AI-based tracking system installed on second offshore rig

Zelim has reported that its AI detection and tracking system, ZOE, has been deployed on a second offshore jack-up rig to support safety efforts in man overboard incidents.

The company said the installation follows more than a year of operation on a North Sea rig managed by an offshore drilling contractor.

ZOE combines onboard camera hardware with machine learning software to detect and monitor man overboard events in real time.

According to Zelim, the system identifies a person entering the water and continuously tracks their position, assisting response teams with faster localisation and retrieval.

The company said the second deployment marks a shift from fixed-site use to more variable environments, where the rig regularly relocates and encounters a range of lighting and sea conditions.

System trained using bespoke maritime visual data

Zelim said ZOE is trained using a proprietary dataset built specifically for maritime detection.

The dataset was first developed in 2020 while Zelim was working on its Guardian unmanned rescue vessel.

Drone-mounted cameras were used to gather images of people in the water under different angles, lighting, and sea conditions.

According to Zelim, these visuals were manually annotated to help teach the system how people appear in non-ideal maritime environments.

The company said the dataset now contains over 7 million labelled images and is one of the most extensive of its kind in maritime safety applications.

Zelim CEO and founder Sam Mayall said: “A person in the water may be wearing dark clothing, face down, partially submerged, or obscured by foam or spray.

“AThese aren’t fixed profiles.

“We had to ensure the system could recognise a human target from a range of angles and distances, under real-world conditions.

“That meant building a dataset that reflected how people actually appear in the water, not how they’re modelled in ideal circumstances.”

Detection module supports other safety operations

Zelim said ZOE integrates with existing onboard systems including navigation and emergency response tools.

The company stated that its detection and alerting process is handled locally, without reliance on remote cloud services.

ZOE is also part of a wider software platform.

Zelim said other modules include Watchkeeper, which identifies navigational hazards, and Shield, which monitors restricted areas for intrusions or unplanned vessel activity.

All modules are based on the same core AI system and have been developed for different safety and operational scenarios.

Mayall said: “If you can reliably detect a person in the water, you can also detect other objects or risks.

“The same system can support navigation, perimeter monitoring, or safety watchkeeping.

“That’s where we see this technology heading – not just detection, but situational understanding.”

System tested in collaboration with US Coast Guard

Zelim has worked with the US Coast Guard to test and validate the AI system.

According to the company, one driver for the partnership was Coast Guard research showing that visual detection by human crews can be inconsistent.

Mayall said: “AI doesn’t fatigue, doesn’t blink, and doesn’t overlook what’s in plain sight.

“That consistency makes it a reliable component in a broader safety system.”

The company said the aim is not to replace human judgement but to support it through faster detection and real-time awareness.

Coast Guard studies cited by Zelim found that visual identification rates for people in the water can fall below 20 percent depending on environmental conditions.

Offshore operators now investing in consistent monitoring

Zelim stated that uptake of the system across offshore platforms reflects a wider shift toward integrated monitoring.

The company said intelligent detection tools are increasingly seen as part of routine safety assurance, not just emergency tools.

Operators are reportedly investing in systems that reduce the time needed to identify and respond to incidents.

Zelim said its long-term development has focused on combining automation with on-site responsiveness.

The company said the use of onboard AI improves both personnel safety and asset awareness, especially in high-risk maritime settings.

Offshore safety enhanced by AI detection system from Zelim: Summary

Zelim has installed its ZOE AI detection system on a second offshore jack-up rig.

The system detects and tracks man overboard incidents in real time.

Its first installation was on a North Sea rig managed by an offshore drilling company.

Zelim said ZOE was designed specifically for maritime safety conditions.

The system uses visual data collected from drone cameras during earlier product development.

This dataset includes over 7 million manually annotated images.

ZOE integrates with onboard infrastructure to function without cloud connectivity.

Other modules, Watchkeeper and Shield, use the same detection engine for separate monitoring tasks.

Zelim collaborated with the US Coast Guard to test the technology.

The company said AI can detect people and hazards with more consistency than human crews.

According to Zelim, detection systems reduce response time and improve operational oversight.

Operators are beginning to adopt such systems as part of their standard safety equipment.