Landing on a moving deck: what offshore drone in a box programs have to solve

Recovery, not launch, sets the mission envelope

Most drone programs spend their design effort on the outbound leg. Offshore, the return is where the engineering lives. A dock on land can assume the pad will be exactly where it was when the aircraft left. A dock bolted to a floating production unit cannot. The platform heaves, the aircraft is moving too, and the last thirty seconds of every mission become a relative positioning problem solved by software instead of by a visual observer standing by. That constraint is why offshore dock programs move slowly, and why they are worth watching when they do. In September, the Journal of Petroleum Technology reported on docked aircraft flying from the operator's Mars floating production unit in the US Gulf, and on an approval that lets the company sign off on its own offshore flights without a visual observer, from any of its Gulf facilities.

What the dock replaces

The inspection task itself is old. After a storm forces a platform to shut in, operators have historically sent helicopters for a visual survey before crews return, which burns flight hours and puts people in the air to answer questions a camera could answer. An autonomous aircraft that flies a repeatable route, holds station near a flare boom and returns high resolution imagery changes the sequence: inspection first, people second. The hardware inside a typical box is fairly standardized now. A docking station, an outward looking camera that inspects the aircraft before launch, and a gas sensor tied to the platform's electrical code requirements. Connectivity runs over satellite plus cellular, and the aircraft launches, captures data, lands and recharges with no pilot on site. Offshore, the box is often sited on a structure where nothing else can be mounted anyway, which is a useful reminder that the dock competes for space with equipment that earns revenue.

The approval path is the real gate Flying beyond visual line of sight in the United States still means a waiver or exemption, and the applicant has to show how the aircraft will detect and avoid other traffic.

The FAA lists operation beyond a pilot's unaided ability to determine position, altitude, attitude and movement as a waivable provision, which is the section most offshore and corridor programs live under. Government auditors reported in February that approved waivers have leaned heavily on ADS-B, with cameras, acoustic sensors and ground radar used as supporting options, and that some stakeholders see added weight as a safety trade off. Self authorization is the point at which a program stops queueing for a regulator's signature on every mission. Three items belong in a program plan before a dock leaves the workshop: - Link loss behaviour over water, where the dock is the only recovery option. Specify what the aircraft does when the command link degrades and how the dock reports a failed recovery. - Glare. Vision based navigation behaves differently over water because sunlight reflects far more strongly than over land, and that affects both inspection imagery and positioning. - Storm sequencing. If the inspection is meant to shorten a return to production after a shut in, the workflow has to be ready to run before the all clear, not after.

Takeaway for onshore teams

Most readers will never site a dock on a platform. The lesson travels anyway. When a dock sits on a structure that moves, floods or gets locked down during an emergency, the recovery plan is part of the mission plan. Write the fallback before the first flight and test it in the worst wind you intend to operate in.