Looking for an artificial intelligence-powered image capture and analysis tool to help with aircraft inspections or damage analysis?
You’re in luck.
Companies you’ve heard of, including manufacturers, and many you haven’t are investing in tools that take pictures and videos of airframes, engines and parts, analyze them and spit out guidance to help technicians. These tools don’t make go/no-go decisions—that’s a large regulatory hurdle. But they help mechanics do their jobs more quickly and thoroughly.
Among the recent and long-awaited breakthroughs is drone inspections. They combine advanced image-capturing with the flexibility of being able to get to places, like the tops of widebody fuselages, more quickly and easily than humans. They also use LiDAR, which can produce high-resolution maps of even microscopic surface damage. And they feed what they capture into AI-powered software for analysis.
Given the flood of offerings out there, what’s left to create that the market will find useful?
Harish Kumar is betting on a combination of existing technology and a bespoke AI engine.
A former automotive program design engineer who spent time developing software in Silicon Valley, Kumar became interested in aviation maintenance when a broken windscreen left him stuck at an airport. After a few years of research, he launched ArcheanX with the goal of giving mechanics conducting exterior inspections a “second set of eyes,” he told Aviation Week.
Again, not a novel concept. But his goal was to connect what those eyes captured to a powerful, accurate analytics tool. ArcheanX built a vision-language model and trained it on aircraft—everything from what a rivet is to what the damage tolerances are on specific models.
The company then developed software that works on off-the-shelf devices, like smartphones and tablets. It captures high-resolution video and LiDAR three-dimensional maps that the AI engine analyzes. A mechanic walks around an aircraft on an inspection and the software captures everything, highlighting possible issues for review.
A handheld tool that goes on existing devices has its benefits. No hardware to buy, and minimal training costs. It can also be used outside of the MRO space, such as if a pilot has a concern during a preflight walk-around.
“We can do landing gear. We can do tire tread. We can do missing rivets,” Kumar said.
While drones have their limitations—they can’t fly anywhere at any time, for starters—their advantages are hard to ignore. A drone can get the same high-quality data in a fraction of the time it takes a human holding a device. Delta Air Lines says its drone-powered general visual inspections take two hours or less, compared to as much as 16 hours for a technician that must access the top of an aircraft as part of the job.
Perhaps the data-capturing element will become less critical than what’s done with it. Kumar is convinced that ArcheanX’s blending of high-resolution video and LiDAR maps and his AI tool will turn prospects into customers.
Several air transport and business aviation MRO providers are evaluating ArcheanX, the company’s founder says, as is one major airline that wants to check brand-new aircraft as soon as they are delivered.
Meanwhile, Kumar says a U.S. military contractor has combined ArcheanX’s core software with its own complementary technology as part of an autonomous aircraft visual inspection pilot program at a major U.S. base. The agreement has not been made public.
Kumar is confident that formal announcements will come soon.
So far, the customers—who are always right—have yet to speak.