Case study · Robotics
The tunnel gets inspected. No one has to walk it.
Railway tunnels have to be inspected regularly — and doing it by hand means people walking kilometres of dark, confined infrastructure. For SNCF, we built a robot that goes in instead.
The challenge
Tunnel inspection is essential, unavoidable — and one of the least automatable jobs in rail infrastructure.
- 01
Inspecting by hand means kilometres on foot through dark, confined infrastructure — slow, costly and hazardous for the people doing it.
- 02
Tunnels are GPS-denied, unlit and unstructured — exactly the conditions where off-the-shelf robots fail.
- 03
A defect report is only useful if engineers know precisely where the defect is. Manual surveys struggle to deliver that precision, kilometre after kilometre.
A robot that understands the tunnel
The robot enters the tunnel and navigates it autonomously — no GPS, no pre-installed infrastructure. Multimodal perception builds a semantic map of the structure as it moves: not just geometry, but an understanding of what it is seeing.
Anomalies are flagged and precisely located on the map for engineers to review — the machine does the survey, people make the judgement calls. All processing runs on-board at the edge, so the robot operates fully autonomously in environments with no connectivity.
The results
Tested on-site in live tunnel environments.
In live tunnel testing, the robot navigates autonomously at 2 m/s, positioning itself with centimetre precision — turning a hazardous manual survey into a supervised, repeatable process.
Have infrastructure no one wants to inspect by hand?
Book a demo and see how an autonomous inspection robot would fit your operations.