An airplane has, for the first time, automatically landed itself after an in-flight emergency, according to the system’s manufacturer.

Two people emerged unscathed from the Beechcraft Super King Air 200 after it stopped on the runway at Rocky Mountain Metropolitan Airport near Denver, according to video posted by emergency responders.

The twin-engine turboprop landed under the control of Garmin’s Autoland system, which the company says is now installed on about 1,700 airplanes. “This was the first use of Autoland from start-to-finish in an actual emergency,” Garmin said in a statement.

  • Demdaru@lemmy.world
    link
    fedilink
    English
    arrow-up
    9
    ·
    6 hours ago

    Poor, poor flight sim folks. Their fantasy of stewardess calling someone able to land in case of problems just got demolished xD

  • Cocodapuf@lemmy.world
    link
    fedilink
    arrow-up
    32
    arrow-down
    1
    ·
    edit-2
    14 hours ago

    It’s disappointing to see so many posts from outraged users railing about the dangers of automation, who clearly haven’t read the article and don’t seem to understand what actually happened.

    This story is actually what it purports to be, it’s an example of a safety system working correctly in an emergency. That’s all it is, not all that exciting to be honest. There was no bug or accidental activation, or AI uprising.

    The only weird part is that the pilots could have landed themselves if they had wanted to, but chose not to. If there were any passengers on board that would have certainly been called a criminally reckless choice. As it was just the pilots though, I suppose it was just bold/dangerous/dumb. (Though on the other hand, they learned more about their aircraft in the process and they tested a critical safety feature, which are both good things.)

    • TubularTittyFrog@lemmy.world
      link
      fedilink
      arrow-up
      4
      ·
      edit-2
      5 hours ago

      any mention of AI/automation triggers people now into thinking it’s evil/wrong. And any rational support of it gets you downvoted.

      People just need something to irrationally hate i guess? It’s funny how every lemmy poster things they are so smart but their responses are overwhelmingy knee-jerk and emotional and often totally disproportionate to the context.

    • UnrepentantAlgebra@lemmy.world
      link
      fedilink
      arrow-up
      1
      ·
      3 hours ago

      The pilots could not turn off the automatic comms which were saying that they were incapacitated. They could turn off the emergency autoland feature as a whole with a press of a button (someone else linked a source for the FAQ).

    • Zak@lemmy.world
      link
      fedilink
      arrow-up
      25
      ·
      edit-2
      19 hours ago

      the pilots COULD NOT TURN THE THING OFF.

      The article does not say that, and I have not seen a source that does. The FAQ says holding the autopilot disconnect button deactivates the system. If two turboprop pilots can’t manage that, then they are incapacitated.

    • Cocodapuf@lemmy.world
      link
      fedilink
      arrow-up
      14
      ·
      edit-2
      19 hours ago

      This almost belongs in the “f* AI” sublemmy.

      Yeah I think that’s going too far.

      Look, the fact that this has never been needed before is a testament to how often all the redundant safety systems work the way they’re supposed to. It’s not like there’s a problem here.

      That said, when the auto-land system was triggered it worked as it was supposed to, landed the plane safely, radio’d its intentions, and nobody was injured. Keep in mind it was triggered during an actual emergency, exactly when it’s supposed to. If the pilots weren’t familiar enough with the safety systems on their plane to disable the auto-land functionality, then they’re functionally ceding control of the plane to the emergency systems, and at that point it’s correct for the automated systems to take over. In this case it seems the pilots decided not to take control back. So really, this was just the safety protocols working as intended.

      I’m glad everyone including the pilots were fine though, that really is the most important thing.

      • piecat@lemmy.world
        link
        fedilink
        arrow-up
        1
        arrow-down
        1
        ·
        14 hours ago

        I almost wonder if it is smarter to have the autoland not be possible to turn off if pilots were for example hypoxic / mentally incapacitated.

        • Cocodapuf@lemmy.world
          link
          fedilink
          arrow-up
          5
          ·
          14 hours ago

          Heh, good luck trying to push an automation technology that a pilot can’t disable. You’re right, there probably are situations where that makes sense, but they would be very rare edge cases. And pilots really do hate being taken out of the control loop. Ideally they really don’t want there to be any computer action that doesn’t have an override (which I think generally makes sense).

        • FuglyDuck@lemmy.world
          link
          fedilink
          English
          arrow-up
          2
          ·
          19 hours ago

          I liked the movie, but the short stories were better.

          I’m especially fond of the one where they developed the robot religion and supplanted the humans that had assembled them because they were more precise/efficient/safe at the task.

          • JetpackJackson@feddit.org
            link
            fedilink
            arrow-up
            1
            ·
            18 hours ago

            I’ve read one of the Asimov collections books, I think it had one of the I robot ones in it. I mostly remember the car retirement one and the one where a man mishears aliens and tells them never to come back and so earth never gets visited again

    • sam@piefed.ca
      link
      fedilink
      English
      arrow-up
      27
      ·
      1 day ago

      and the pilots “made the decision to leave the system engaged,” Townsley said.

    • plantfanatic@sh.itjust.works
      link
      fedilink
      arrow-up
      18
      arrow-down
      3
      ·
      1 day ago

      So it still landed automatically though. That’s the entire point of the article. Pilots didn’t control it, the fact that it wasn’t a real emergency is entirely moot.

      • solrize@lemmy.ml
        link
        fedilink
        arrow-up
        25
        arrow-down
        3
        ·
        1 day ago

        Well no, the whole point of the news event was the supposedly real emergency. Autoland has worked in testing and training for yearrs. This time it was basically credited with saving humans, but it didn’t do that. Instead it surfaced a bug in its own progamming, it sounds like.

        • Cocodapuf@lemmy.world
          link
          fedilink
          arrow-up
          11
          ·
          20 hours ago

          What bug? It worked as it was supposed to.

          There was a loss of cabin pressure (a real emergency) and it decided the pilots were unresponsive. The pilots could have taken control back, but by not doing that, they were functionally unresponsive. I don’t see a bug here.

        • LifeInMultipleChoice@lemmy.world
          link
          fedilink
          arrow-up
          4
          arrow-down
          1
          ·
          21 hours ago

          The first automatic landing of a commercial airliner was apparently in 1965. So yeah it was just the entire process for this individual emergency system by this company being call a first as well. 60 years ago apparently we succeeded, which makes sense, 4 years before we landed on the moon, believe that was 69’

      • gravitas_deficiency@sh.itjust.works
        link
        fedilink
        English
        arrow-up
        15
        arrow-down
        7
        ·
        edit-2
        10 minutes ago

        This falls in the same category as the MCAS crashes. An automated system wrested control from the pilots, and the system evidently could not be disabled. That is the exact scenario that caused the MCAS crashes. While the outcome here was much more positive than “it crashed”, an automated system refusing to allow pilots to exert manual control is Not Okay. The positive outcome is incidental; the refusal of the automation to allow resumption of manual control is the core issue. That is a flaw that should ground the system until a bug fix is deployed. My dogmatic opinion on this comes from having worked in aerospace previously - this isn’t ivory tower speculation/assertion.

        Edit: rtfa lol - the pilots left it engaged intentionally. There’s nothing wrong here - this wasn’t a control authority override.

        • atzanteol@sh.itjust.works
          link
          fedilink
          English
          arrow-up
          3
          ·
          6 hours ago

          pany CEO Chris Townsley in a statement. Autoland “automatically engaged exactly as designed when the cabin altitude exceeded the prescribed safe levels” and Townsley said.

          the pilots “made the decision to leave the system engaged,”

          the pilots made the decision

        • Cocodapuf@lemmy.world
          link
          fedilink
          arrow-up
          10
          ·
          edit-2
          19 hours ago

          This falls in the same category as the MCAS crashes. An automated system wrested control from the pilots, and the system evidently could not be disabled. That is the exact scenario that caused the MCAS crashes.

          The only thing though, is that is absolutely not what happened here, at all.

          Here’s a quote from the (quite short) article:

          Autoland “automatically engaged exactly as designed when the cabin altitude exceeded the prescribed safe levels” and the pilots “made the decision to leave the system engaged,” Townsley said

          The pilots saw the warnings and presumably could have taken over by doing just about anything, but with no passengers to put in jeopardy, they decided to sit back and see what happened.

          There is no flaw here, no bug, this is not a problem. This is what happens when everything works.

          Edit: Well, not everything was working, that beechcraft does have to have its pressurization system checked out for sure.

        • ThisGuyThat@lemmy.world
          link
          fedilink
          arrow-up
          8
          ·
          1 day ago

          Yea not every airport has a tower. The Garmin system cannot detect obstructions. It announces an intent to land and lands.

          • brygphilomena@lemmy.dbzer0.com
            link
            fedilink
            English
            arrow-up
            1
            ·
            8 hours ago

            If it finds an airport with an ILS runway, and it has even remotely up to date aviation charts with minimum altitudes what obstructions does it need to avoid?

            Uncontrolled airports probably aren’t where this system is programmed to fly into in an emergency. Not only would that be more difficult to program, but I suspect there wouldn’t be emergency crews to assist with whatever the emergency was either.

          • frongt@lemmy.zip
            link
            fedilink
            arrow-up
            1
            ·
            23 hours ago

            Does it only choose suitable airports? Like ones where they have a system to support automatic landings? I would think those should have ATC

          • sramder@lemmy.world
            link
            fedilink
            English
            arrow-up
            1
            arrow-down
            2
            ·
            23 hours ago

            So… my $1400 DJI Avaya2 has a more sophisticated automatic landing system? Yeah… lots of stuff just started making sense 🤣

              • sramder@lemmy.world
                link
                fedilink
                English
                arrow-up
                2
                arrow-down
                3
                ·
                23 hours ago

                And presumably lots of other expensive acronyms that I don’t care about, because I’ll never own a plane 🤣

                • Riskable@programming.dev
                  link
                  fedilink
                  English
                  arrow-up
                  6
                  ·
                  edit-2
                  23 hours ago

                  Think the acronyms are bad now? Wait until your plane is in the cloud!

                  “Plane pod bay doors open unexpectedly! Engaging emergency AWS FaaS! AWS unavailable! Engaging GCP FaaS! GCP unavailable! Local CDN AZ cannot be found! No MSP available! Engaging SRE! Blaming DNS!”

                  Pilot: “We haven’t even taken off yet! Also, I told you to open the doors, you dumb AI!”

                  “Good catch! Would you like me to continue blaming DNS?”

        • plantfanatic@sh.itjust.works
          link
          fedilink
          arrow-up
          3
          ·
          edit-2
          8 hours ago

          The cabin lost pressure, it activated because an emergency happened. Then the pilots decided to leave it on to see if it worked during an actual emergency situation.

          The very short article covered these details lol.

  • nondescripthandle@lemmy.dbzer0.com
    link
    fedilink
    arrow-up
    7
    arrow-down
    1
    ·
    edit-2
    19 hours ago

    Honestly surprised this is the first time, this isn’t that new of a feature when it comes to airplanes, even in commercial jets it predates chatgpt by years. Thanks to Instrument Landing Systems, which have been around an even longer time, pilots can land using solely the instruments in the cockpit without ever looking out the window. It’s not easy to automate landing and many pilots wouldn’t trust it except as a very last resort but their automation problem does avoid some challenges an pitfalls autonomous cars run into.

    • UnrepentantAlgebra@lemmy.world
      link
      fedilink
      arrow-up
      2
      ·
      4 hours ago

      Non-emergency autoland like you describe had been around for a long time. But you have to manually pick your flight plan and approach, and at least point the aircraft in the general direction of the flight plan so that the AP will engage and handle the rest.

      This is a totally different beast because the emergency autoland picked the route and approach, managed the AFCS and comms, and then did the normal autoland stuff after all that. Garmin’s emergency autoland only came out a few years ago.

    • Cocodapuf@lemmy.world
      link
      fedilink
      arrow-up
      4
      ·
      19 hours ago

      So did the space shuttle most of the time. For the first few years the shuttle was piloted manually, but that changed at some point and I believe the majority of flights were landed automatically.

      The buran on the other hand only actually flew at all once.