Slashdot Mirror
EasyJet Turning To Drones For Aircraft Inspections
itwbennett writes: Would you trust your aircraft inspection to a drone? Budget airline easyJet is testing just such a system, aimed at reducing the amount of time an aircraft is out of service. Instead of having humans perform on-site visual inspections, the drone will "fly around an aircraft snapping images, which will then be fed to engineers for analysis."
Rather than standing around making a judgment, take some pictures and use them to decide later. This could be done with a handheld camera as well. Guess you make the news when you mount it on a drone. Since the equipment is expensive, keeping it in service a higher percentage of the time makes sense. Unless the drone can get the pictures faster than a person, it's really superfluous in this context. It may be cheaper, though, because you don't need to manage the photograph takers.
Now they'll just blame the engineers looking at the photos.
With sufficient resolution (as TFA points out) and the ability to finely control the drones this could be a good thing since you could hover near and area and get a closer look; and then do a normal inspection if you still have questions.
I'm a consultant - I convert gibberish into cash-flow.
FTA: "Currently, such inspections are done visually and require an engineer to get up above the aircraft and around its exterior. That requires a platform and takes valuable time..." I predict that, with reduced budget and manpower for actual inspections and political pressure to certify planes as "passing" when they aren't, easyJet will have a significant mechanical failure within five years. It will might just be spectacular enough to make the news.
If I can build a rudimentary 3D scanner with decent resolution out of a cheap laser pointer, a wine glass, and a 480px resolution webcam, surely a fleet of expensive drones sporting modern, HD cameras could do the same thing a few orders of magnitude more efficiently.
An enigma, wrapped in a riddle, shrouded in bacon and cheese
in multiple light spectrums to give more information, maybe. Some cracks can't be seen visually but under infrared or UV they can be seen.
Nobody's leaving aircraft inspection to the drones. They're just a tool like any other. As long as there's no question that the images reviewed are of the right aircraft (no spoofing, please), I think it's really no different from using a 20 meter selfie stick to take the pictures.
What do you mean they cut the power? How can they cut the power, man? They're animals!
... fly the drone within 5 miles of the airport.
It little behooves the best of us to comment on the rest of us.
Would you trust your aircraft inspection to a drone?
Never trust a drone, man.
I learned that the hard way. I loaned one $20 and it just flew away with my money.
and on.
A quadcopter can fly a route that is preprogrammed for each type of airframe. As long as it is able to index off a defined point for each aircraft, it should be able to fly to each key inspection area and take exactly the same photo from exactly the same angle every time. Those photos can then be processed by computer to compare differences between thousands of nearly identical samples to determine a variance metric, which can help a human worker prioritize the images for review and referral to a ground inspector. Otherwise the computer inspector is going to be looking at lots of nearly identical pictures so any difference should stand out like a sore thumb.
fly the drone within 5 miles of the airport without authorization from the tower.
Failures due to lack of inspection because the inspection activity is a PITA is a hard nut for engineers to crack. Big aircraft have hard-to-reach spots that require ladders and man-lifts to access. It just takes one lazy technician to skip an inspection to miss a flaw that leads to a failure.
This could be a good idea if it facilitates inspection of parts of the aircraft that are normally difficult to access (e.g. boroscopes for engine inspections is a similar idea). The top of a T-tail aircraft that would require a man-lift and some time to inspect something that is quickly and plainly visible once you are in position - this would be a perfect application of a quadcopter with a camera (fuck off with the "drone" meme, please. It's an RC copter with a GoPro).
If it makes it easier (and maybe fun) to do the right thing, cool. It sounds childish, but if you can make a critical job easy and fun, you increase the chances that the job gets done enormously. Some (not many) engineers think to design things in this way.
Left MS Windows for Linux Mint and never looked back!
Vote for Bernie in 2016!
Am I the only one annoyed by any radio controlled vehicle now receiving the drone label? I mean radio controlled vehicles have been around forever. If they cannot fly fully autonomously then stop calling them drones. Who said these aircraft weren't going to be radio controlled and operate autonomously? Drones as a default label sucks.
Now these are drones: http://en.wikipedia.org/wiki/The_Culture#Drones
RIP Banks
Assuming the cameras are sufficient resolution to capture the small defects before they become larger defects (or cause a failure) and assuming there is still a way to a human to get up there and look at something if the photos aren't clear enough and there are still questions, I see no problem with this.
Drone sucked into the jet's intake?
Don't be apathetic. Procrastinate!
More job reduction, lower performance by drones.
"If any question why we died, Tell them because our fathers lied."
I was under the impression that lots of stuff that needed inspecting lay behind some sort of access panel. Or in some location not easy to get to. I don't think a drone will be much help for this sort of thing.
On the other hand, prior to flight, it is necessary to do a 'walk around' of the aircraft looking for problems like fluid leaks, access covers inadvertently left open, locking pins left in place and other similar items. Customarily, this is done by the flight crew. But a drone might be able to get a better look than a person on the ground. This could catch goofs, like the duct tape over the static ports that resulted in the Aeroperu Flight 603 crash.
Have gnu, will travel.
Inspecting the tail fins, and the top of the fuselage is far easier, quicker, and cheaper with a drone.
I agree that it might be easier, quicker and cheaper with a drone. However I don't really care. As a passenger I'm far more interested in whether it is just as effective as spotting problems as the human eyeballs it replaces. On the plus side images can be zoomed and you might see more detail than a human eye. On the downside the image is probably not going to be 3D and it sounds like the person taking the pictures with the drone will not be the engineer who inspects them.
Maintenance caused the failure, but it was unquestionably pilot error that caused the crash.
Actually not according to the linked article. The pilots followed the correct procedure which was a slow climb with flaps open but the engine falling off the wing severed the hydraulic lines and caused a partial power failure which meant that the slats retracted on the one wing and the warning indicators both for stall and asymmetric slats did not work. The crash might have been avoidable given hindsight but I would not call it pilot error by any stretch of the imagination and again according to the Wikipedia article neither did the NTSB.
The drones are cheap and the engineers in question will be a massive indian poorly educated wannabe slave army of picture lookers who will not find what they are suppost to find.
What a poor solution for making airplanes save. No instead we reduce engineers at the side in US or Europe and outsource everything to someone who is barely able to distinguish a plating texture from a drilling hole.