Slashdot is powered by your submissions, so send in your scoop

 



Forgot your password?
typodupeerror
×
Social Networks The Internet

Facebook's Solar-Powered Drone Under Investigation After 'Accident' (theguardian.com) 45

Facebook has hit a hitch in its plans to use a solar-powered unmanned drone to provide internet access to developing nations, after it was revealed the American National Transportation Safety Board (NTSB) has opened an investigation into an accident on the drone's first test flight in June. From a report on The Guardian:At the time, Facebook described the flight as "successful": the drone, called Aquila, stayed aloft for 96 minutes, three times the planned duration. "We have a lot of work ahead of us," Jay Parikh, Facebook's head of engineering and infrastructure, wrote when Facebook revealed the test flight, in late July. "In our next tests, we will fly Aquila faster, higher and longer, eventually taking it above 60,000 feet." In a second, more technical, blogpost published that same day, Facebook's Martin Luis Gomez and Andrew Cox acknowledged the failure in passing. "Our first flight lasted three times longer than the minimum mission length, so we were able to gather data on how the structure and autopilot responded under a range of real-world conditions to help verify these predictions," they wrote.Reporter Casey Newton mentioned on The Verge that at the time, Facebook had led them believe that everything was alright, and there were no hiccups.
This discussion has been archived. No new comments can be posted.

Facebook's Solar-Powered Drone Under Investigation After 'Accident'

Comments Filter:
  • missing quote (Score:5, Informative)

    by sacrilicious ( 316896 ) <qbgfynfu.opt@recursor.net> on Tuesday November 22, 2016 @10:47AM (#53338747) Homepage

    In a second, more technical, blogpost published that same day, Facebook's Martin Luis Gomez and Andrew Cox acknowledged the failure in passing. "Our first flight lasted three times longer than the minimum mission length, so we were able to gather data on how the structure and autopilot responded under a range of real-world conditions to help verify these predictions," they wrote.

    I read that three times trying to figure out whether the "in passing" mention of failure was so subtle that I was missing it. Nope, the editors simply left out the actually relevant quote:

    “We are still analysing the results of the extended test, including a structural failure we experienced just before landing. We hope to share more details on this and other structural tests in the future,” Cox and Gomez added.

    • Wow, carbon copy experience here.

      Thanks for the clarification.

    • by asylumx ( 881307 )
      Yup, same here. Honestly, it's not clear to me what the news is in this article.
    • "an accident just before landing" - sounds like it either hit a power line or some other obstacle - or a bird took it out. If they can take down big airlines, what chance does a drone have?

      Or maybe because they kept it up 3x as long as it was designed, something let go that wouldn't have on a shorter flight. We just don't know, and the article doesn't say.

      Was the flight a success? Sure - it exceeded it's planned mission goals. Not bad for a drone with a 130' wingspan. And it also found a possible design o

      • Could just be an old fashioned tip stall. Not enough information.

        Flying wings have poor yaw control. At landing it's near stall speed. Little side wind, drone yaws, inside wingtip stalls, roll, smash. Long wingspans makes it worse.

        Pure speculation.

      • Or maybe because they kept it up 3x as long as it was designed, something let go that wouldn't have on a shorter flight.

        Come on, this isn't a rocket under tremendous stresses or a Mars rover in a harsh environment, it's a bloody drone, albeit a high altitude one. Experimental model airplanes cobbled together by hobbyists over the weekend from balsa wood and shrink foil last a hell of a lot longer than 96 minutes without maintenance (even though they don't stay aloft that long in a single flight). If these guys cannot design a prototype that flies for that long, then I hope the regulators will take a very long and hard look

        • Re:missing quote (Score:4, Insightful)

          by LenE ( 29922 ) on Tuesday November 22, 2016 @02:47PM (#53340667) Homepage
          OK, Internet super-engineer. You clearly do not know anything about aircraft or spacecraft structures.

          Materials used for small drones have a very difficult time scaling up to very large lightweight structures. Care to expound on the loads and stresses experienced by a 144 foot wingspan wing, that weighs only 900 lbs? This drone was built to be light and barely strong enough. That was its design point. Have you shown the world that you can do better and can credibly criticize their design?

          Extremely high aspect ratio wings, like this one, just don't want to quit flying. It is a real challenge to bring it down onto the ground. The pilots possibly exceeded design Vne trying to get it down. This exponentially increases drag forces on the airframe. Ground-effect makes landing even more difficult, and with a 144 foot wingspan, ground-effect starts at ~72 feet above ground.

          Add to this the non-homogeneous nature of gusting winds in proximity to the ground, and it is not inconceivable that design limits were exceeded by a fluke of nature. There is not enough public information of what weather conditions existed during landing, but sudden adverse conditions during landing are not unprecedented [wikipedia.org]. Did Lockheed's engineers under-design the structure on the L-1011?

          Disclaimer - I work for an aerospace prototyping firm (not related to Facebook), and have worked on things that cover the conceivable gamut of Reynolds numbers that can be "flown". Although I have not worked on Facebook's drone, I have an intimate knowledge of the modern materials used in this air vehicle. My opinions expressed here are my own.

          • So it's normal to design prototype aircraft that are expected to stay up for 30 minutes before failing but not 90? And if they are designed that way, is it normal to fly them until they break anyway? Because that was my point, as GP suggested that the mechanical failure might be due to flying the craft a bit longer than planned. That just seems silly and unlikely.
  • Comment removed (Score:3, Interesting)

    by account_deleted ( 4530225 ) on Tuesday November 22, 2016 @10:53AM (#53338787)
    Comment removed based on user account deletion
    • by Anonymous Coward

      2. How will you fly drones in foreign airspace without the consent of a host country?

      3. How do you justify brining internet to countries like Malawi, Zamibia, and Angola when a grain shipment or food programme would do far more to improve the lives of these people than another American drone?

      4. Egypt and South Africa already have high-speed internet available to the general public. did you forget this? or are you just trying to erode public investment in open and neutral networks?

      Wow... okay, I'll bite because your post is just idiotic enough to warrant a response.

      Where is it that you think testing a drone in American airspace equates to invading another country? Of course a private company will need to get permission from the relevant agency in India. And getting that permission will likely cost quite a bit of money.

      And why does anyone have to justify providing a service rather than another handout of food? If they can't even feed themselves without freeloading off of grain ship

    • by Anonymous Coward

      Food programs don't work in those countries because their government is fucked up. There is no way to help them without fixing their government first.

    • Re. point 3: teach a man how to fish and all that. That's what the internet is for, and even crappy limited access through feature phones has already improved the lives of countless people in developing nations, far more than any one shipment of grain would have. Access to information on weather and prices, access to markets, mobile payments, all these have enabled people to improve productivity, operate in new markets, and start new businesses.

      Drones might be a good solution to deliver Internet to suc
    • 1. Because bowing to the will of a government that is not necessarily always acting in the interests of the people has never gone poorly?
      2. I'm sure they haven't considered this at all.
      3. Food programs do absolutely fuck all. Information, knowledge, and the ability to communicate on the other hand change lives.
      4. Hahhahahaha yeah so does North Korea.

  • From Facebook's perspective, the best thing about testing "developing countries" is probably that dropping a drone on someone's head from 60,000 feet will mean, at worst, they're out a three goats and a pair of chickens. (See also: drug makers and name-a-country.)
    • by DarkOx ( 621550 )

      Well India probably isn't the place than. While the availability of justice to the average Indian might still leave a lot to be desired, it is a nation with a strong legal frame work (when used) and international pull to go after a company like facebook.

      Where FB is concerned deep pockets will get attention anytime there is blood on the water. They would be much better off conducting dangerous experiments in one of the many failed states, where they could pretty much count on being able to fire up the corp

    • Oh no... Once the villagers figure out WHO was responsible and how much $$ they have, you can bet the army of ambulance chasing local lawyers will be out in force making sure Facebook pays though the legal system.... Unless they have bribed the proper local authorities to skirt the legal system.

      Either way, they will pay...

    • by ShaunC ( 203807 )

      Except they aren't testing (solely) in developing countries; this test and whatever anomaly ensued took place in Arizona. Not exactly the population center of the US, granted, but the tests are being done here at home.

      Facebook isn't alone in this regard, either, as Google has its own fleet of experimental drones. N749G flew over my house in the Memphis suburbs last Monday night enroute from KSIK to KOLV. The FAA says it's an "Ashfloyd Hummingbird," whose manufacturer has essentially zero public presence, bu

  • by 0xG ( 712423 )
    This is interesting how?
    • This is interesting how?

      Facebook crashed and burned and now the NTSB is investigating? How's that not interesting? An internet company catches the attention of the National TRANSPORTATION and SAFETY board seems like interesting news to me.

  • "A horrible Facebook Solar drone accident"

    "In shocking news today, Facebooks solar powered drone sucked up all available sunlight for a period of about 96 minutes..."

  • As already pointed out (+5 by sacrilicious), the TFS doesn't make sense. Why not update the summary with the missing line of the quote and include the disclaimer "EDIT: summary updated with full relevant quote".

    Mistakes are human, but fixing them is an act of kindness to the readers that don't hit the site in the first wave.

BLISS is ignorance.

Working...