HyLight raises €13 million for AI inspections
HyLight has announced €13 million in funding to expand its autonomous airship inspection business. The French company combines aircraft, sensors and AI software to help operators assess power lines, pipelines and railways.
The financing involves Rubio Impact Ventures, DRW and Teampact Ventures, alongside existing investors Y Combinator and Ring Capital. HyLight plans to increase its fleet from four aircraft to 15 operating simultaneously in 2027, while expanding production and automated data analysis.
Why an airship suits inspection
The design separates lift from propulsion. In an October 5 interview with the Île de France regional government, chief executive Martin Bocken explained that helium keeps the HyLighter airborne while a hydrogen fuel cell provides energy for movement. He said the aircraft can remain aloft for up to ten hours. Its roughly 35 kilometre per hour cruising speed is intended to support detailed observation.
HyLight describes a payload capacity of up to ten kilograms for equipment including LiDAR, optical sensors and gas leak detectors. That gives the aircraft different ways to examine the same route. Its product website also describes remote communications through radio, mobile networks and satellite connections for operations beyond the pilot’s visual line of sight.
Turning flights into maintenance decisions
Bocken described a service that begins with transporting an aircraft by road to the network being inspected. It follows an automatic flight plan with sensors chosen for the mission. The company then analyses the collected data and supplies maintenance teams with a diagnosis covering problems such as damaged infrastructure, threatening vegetation or leaks.
The software layer is central to that workflow. HyLight presents a dashboard with AI assisted anomaly detection and three dimensional reconstruction, linking aerial collection to a view of infrastructure condition. The value for an operator is in knowing where to investigate or carry out maintenance, rather than receiving an unfiltered collection of images.
Evidence from a railway mission
In a recent company update, HyLight described inspecting 50 kilometres of the South Europe Atlantic high speed railway for operator LISEA in three hours using one aircraft. It said the mission collected photographs and LiDAR data together, producing a three dimensional representation alongside the visual inspection. Those figures are HyLight’s account of that mission, rather than an independently verified performance benchmark.
Scaling the service
Infrastructure customers have long decision cycles. In the interview, Bocken also described work with French and European authorities on permissions for larger operations beyond visual line of sight. A bigger fleet is only part of expanding the service.
The aircraft is one part of a service that turns observations into maintenance decisions. HyLight is funding expansion of the complete workflow, while the larger fleet remains a future target.
Our report on ANYbotics Shift follows a related problem of turning robot inspection findings into maintenance work.
