EyeROV: robots that dive so divers don't have to
A Kochi startup's underwater robots inspect dams, ship hulls and pipelines for industry, then survey coral reefs in the Andaman and Lakshadweep islands for marine scientists, using the same AI vision layer.
I came across this while browsing India's NITI Aayog frontier-tech showcase, in a category I wasn't expecting: underwater robots that also moonlight as coral reef researchers.
The company is EyeROV, founded in Kochi in 2017 by two IIT graduates, Johns T Mathai and Kannappa Palaniappan, who arrived at the same problem from opposite ends. Johns came from robotics at GreyOrange; Kannappa spent time on research vessels, watching ship hull inspections stall for days because they needed a diver who wasn't available. Their answer was to build robots for the job instead: TUNA, a compact 4K inspection robot rated to 300 metres, SAGARA, a bigger six-degrees-of-freedom platform with dual zoom cameras, a pipeline crawler called TSROV that can travel 10 kilometres inside a tunnel, and a mapping boat, iBOAT ALPHA.
Here's the clever part. The hardware is really a delivery vehicle for EyeROV's own AI layer, EVAP, which processes the live underwater feed to cut through sediment and low light, then automatically flags corrosion, cracks and leaks instead of leaving an engineer squinting at murky footage. Point that same defect-detection pipeline at a different subject, and it becomes a research tool: EyeROV's robots now survey coral reefs in the Andaman and Lakshadweep islands, letting marine biologists track bleaching and biodiversity without the repeated dives that stress fragile reefs.
It's a neat case of one hard engineering problem, seeing clearly and reliably underwater, unlocking two very different markets: industrial inspection that would otherwise depend on divers and weather windows, and conservation science that benefits from being left alone. EyeROV now operates in five countries out of a small Kochi team. Nothing about it is flashy, but it's the kind of infrastructure-grade AI that quietly replaces risk with data, and I like that it started from two people comparing notes on the same annoying problem.
