Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
AI

DARPA Developing the Ultimate Auto-Pilot Software 75

Posted by timothy
from the have-they-not-seen-airplanes-1-or-2? dept.
coondoggie (973519) writes "Call it the ultimate auto-pilot — an automated system that can help take care of all phases of aircraft flight-even perhaps helping pilots overcome system failures in-flight. The Defense Advanced Research Projects Agency (DARPA) will in May detail a new program called Aircrew Labor In-Cockpit Automation System (ALIAS) that would build upon what the agency called the considerable advances that have been made in aircraft automation systems over the past 50 years, as well as the advances made in remotely piloted aircraft automation, to help reduce pilot workload, augment mission performance and improve aircraft safety."
This discussion has been archived. No new comments can be posted.

DARPA Developing the Ultimate Auto-Pilot Software

Comments Filter:
  • Thats what was flying Flight 370 !!
    • by aberglas (991072)
      It was running malware written by the NSA that was accidentally released into the wild...
    • Thats what was flying Flight 370 !!

      Shall we call it Flight S/370, then?

  • The vast majority of heavy aircraft losses are due to pilots. They are by far the weakest link in the chain. Whether they do something extremely stupid such as hold back the stick in a stall (Air France) or malicious (Malasian Air) the result is the same. Perfectly good aircraft destroyed.

    So engineering them out of the cockpit is the next step. Computers have been able to fly entire flights for decades. Expert systems already out perform people at diagnostics. The pilot is just a redundant point of

    • by michelcolman (1208008) on Saturday April 19, 2014 @05:22AM (#46793869)

      This is one of the most often repeated misunderstandings in aviation: the vast majority of crashes is caused by pilots, so we should replace them with automation since that's much more reliable. Errr... no, not by a long shot.

      The vast majority of crashes is due to pilot error because the vast majority of possible crashes due to equipment failures are prevented by the pilots. I am a pilot, have never been in a crash, but have had several autopilot and other failures where, if we had not intervened, the aircraft would have crashed. But of course, all those possible crashes due to equipment failures don't make it into the statistics because no actual crash occurred. It's merely a note in the company's safety magazine for crews (along with dozens of others each month). So when an aircraft does crash (even if it's due to equipment failure), it's usually still considered the pilots' fault, and correctly so, because they should have been able to prevent it.

      Take the Turkish Airlines that crashed in Amsterdam. Due to a radio altimeter failure during an automatic approach, the aircraft thought it was directly above the runway and pulled the throttles back, while in fact it was still several hundred feet above the ground. Most crews would have seen the speed decreasing (and indeed, this kind of incident had happened many times before to other crews without causing a crash) but this crew reacted much too late and "caused" the airplane to crash.

      Or take the Air France that crashed after the pitot tubes froze up. The automation actually failed so the pilots had to take over. Without pilots, the airplane would have crashed anyway. And here, too, this kind of incident had already happened to other crews multiple times, but each time the crew had handled the situation correctly (even though it was not something that was trained in the simulator or accurately described in the procedures). This time the crew did not handle it correctly, in part because they were confused by conflicting warning messages from the airplane's systems telling them the plane was overspeeding and stalling at the same time. They even got aural warnings when they started to, temporarily, apply the exact correction they needed to meke. The automation was not helping them, but actually working against them and telling them they were wrong when they were, in fact, right.

      If you want to have an idea of how reliable automation is, just look at the number of military drones that have crashed so far. Their mission couldn't be simpler: take off, fly over some area, come back and land. They only fly in relatively nice weather, there are vaslty less drones than passenger aircraft, yet there are many more drone crashes than passenger aircraft crashes.

      It's certainly a good thing that Darpa is trying to make aircraft automation more reliable, but right now pilots are still by far the most important asset for the safety of an airplane.

      • This is one of the most often repeated misunderstandings in aviation: the vast majority of crashes is caused by pilots, so we should replace them with automation since that's much more reliable. Errr... no, not by a long shot.

        Many good examples snipped

        It's certainly a good thing that Darpa is trying to make aircraft automation more reliable, but right now pilots are still by far the most important asset for the safety of an airplane.

        You make a number of good points. Automation is great when everything is going well; the biggest problem then with automated systems is boredom or an unwillingness to use the system because they didn't chose that career to simply sit back and watch gauges. Automated systems, when everything is working as planned can often do better than a human imply because they can take many more inputs and respond to them than a human.

        However, when sensors start sending conflicting information automated systems start hav

      • by Rich0 (548339)

        Good points. If humans were to be taken out of the loop obviously it will be necessary to change how the automation works, sometimes substantially. Anything that causes an autopilot disconnect, for example, obviously has to be redesigned (well, aside from pilot-triggered disconnects). There may also need to be an increase in redundancies as well so that the plane can remain fully automation even with failures. The algorithms also have to be designed to better handle a lack of sensor input, since all the

      • by w3woody (44457)

        It's also worth remembering that, as pilots are the last line of defense when equipment starts malfunctioning, when a crash occurs the NTSB often sights "pilot error" for failing to maintain control during an equipment malfunction. It may be that the root cause of the failure was equipment malfunction, but unless a wing falls off the aircraft, the pilot is expected to maintain control of the airplane through all equipment malfunctions.

        Anytime I hear about someone talking about making a better autopilot or s

        • by jbwolfe (241413)

          Autopilots often make things more difficult for a pilot because, in some circumstances, the autopilot simply adds a new workload layer that can sometimes interfere with operations.

          That is exactly how we are trained with regard to the use of automation: If its increasing your workload, turn it off. We are encouraged to occasionally fly not only without the autopilot, but also without flight directors and autothrust off. The idea being to maintain proficiency.

        • by Rich0 (548339)

          The whole concept is to get the pilot out of the cockpit entirely - they wouldn't be managing the autopilot - it would be managing itself.

          Plus, if updates to the plane's mission were made while in-flight, it would be done by a team working from desks. They would have time to properly follow procedures, and wouldn't have long stretches of idle time (it would be like a call center - they just move from one plane to the next).

          This would require changes to how we manage planes, ATC, etc in general. However, I

      • by jbwolfe (241413)

        Take the Turkish Airlines that crashed in Amsterdam.

        This incident has some similarities to the Asiana crash in SFO. In both cases, pilots failed to recognize FMA's (flight mode annunciation). In Schipol, the autothrust had changed to retard mode (used during the flare) which allows the airplane to slow below ref speed and land. In SFO, they may have disarmed the autothrust instead of disconnected it, the difference being that they bypassed the low speed wakeup function of the autothrust which prevents low energy conditions.

        In both cases, pilots lacked under

      • by Rich0 (548339)

        If you want to have an idea of how reliable automation is, just look at the number of military drones that have crashed so far. Their mission couldn't be simpler: take off, fly over some area, come back and land. They only fly in relatively nice weather, there are vaslty less drones than passenger aircraft, yet there are many more drone crashes than passenger aircraft crashes.

        I hate to reply twice, but I was giving this some thought. How many of these crashes are the result of faulty automation?

        Virtually all airliners have redundant everything, especially for critical components like engines. Drones usually don't have these things - if the predator's single engine fails, then it crashes. Sure, they could put two engines on them, but the cost of doing that is higher than the cost of buying a new one from time to time. They're designed to be expendable.

        A computer-piloted passe

      • > The vast majority of crashes is due to pilot error because the vast majority of possible crashes due to equipment failures are prevented by the pilots.

        Before automation, the vast majority of crashes were caused by pilot errors or pilots failing to respond properly to emergencies due to being too tired from flying the plane. Automation lets pilots save most of their energy for when it counts instead of tiring themselves out with routine flight procedures like keeping their hands on the control column fo

    • by jbwolfe (241413)

      If you know any pilots put this to them and watch the response ;)

      WTF does that mean? Am I supposed to react with giddy agreement that my profession is pointless? Using your logic, humans need never do anything that can be automated- surgery, programming, procreation...

      No artificial intelligence can replace the versatility of the human mind. Pilots are there for the ability to make decisions under widely varying conditions. The automation is there to lessen the work load.

      The vast majority of heavy aircraft losses are due to pilots.

      Yeah, and when your idea of the pilot-less cockpit is attained it will be "The vast majority of heavy

    • I think that's what they call it when there is no evidence of a mechanical failure and cannot prove that there was an error on the part of the flight crew. The assumption is generally pilot error of some kind. (assumption)
  • by JustOK (667959)
    Why would they use an auto-pilot for an airplane? Shouldn't they use a plane-pilot for planes?
    • Re: (Score:1, Redundant)

      by CeasedCaring (1527717)

      Why would they use an auto-pilot for an airplane? Shouldn't they use a plane-pilot for planes?

      Auto refers to automatic not automobile.

    • by paiute (550198)

      Why would they use an auto-pilot for an airplane? Shouldn't they use a plane-pilot for planes?

      Plane-pilots don't have that emergency reinflation tube just below the belt buckle.

  • Having said that, many recent aircraft failures have been caused by the crew and I think fully automatic airliners should be looked at. Or at least keep a hostie around to blow the thing up.

    • by mmell (832646)
      The problem is that a human somewhere has to set up the automation. Even with the best designed automation, a human is involved somewhere. I'd like a human (or a computer that is at least as intelligent as a human) on site to mediate that risk. That means a flight crew aboard all commercial jets. I'm willing to accept the higher cost associated with that, in return for not putting absolute faith in a thing designed by the hand and eye of Man.
      • Yeah but the problem with human drivers/pilots is that they send SMS messages, update facebook, phone their girlfriends or go crazy and decide to kill their passengers. The engineers who build the vehicles do all those things too, but in an environment where their work can be checked and peer reviewed.

        The best mass transit system I have seen is in Kuala Lumpur. It has no drivers. It is faster and more reliable than any other system I have seen.

  • Could this tech finally solve the "everyone would need a pilots license" problem of consumer level flying cars? Maybe cars could be developed that rely on the person inside the to drive on roads but as soon as liftoff is initiated an auto-pilot like this one DARPA is making could take over completely removing the human factor from the flying hunk of metal.
    Not saying it's imminent but perhaps this is a step in the direction of ubiquitous personal flying vehicles that could solve a lot of transportation pr
  • I've long thought that it would be good if cargo ships were automated and/or remote controlled. Piloting cargo ships ought to be relatively easy compared to remote piloting drones in combat.

    - Crew is valued higher than cargo, so piracy relies a lot on kidnapping crews for ransom. No crew, no crew ransom. That would change the bargaining balance.

    - Also, if crews can't manually override the automatic/remote control, then they have no control over the ships course. That would make piracy even more difficul

    • I've long thought that it would be good if cargo ships were automated and/or remote controlled. Piloting cargo ships ought to be relatively easy compared to remote piloting drones in combat.

      Many are. You set Iron Mike on a course and speed and he follows it. You still, however, need someone on watch to watch for and deal with the inevitable unexpected situation.

    • by Rich0 (548339)

      Docking and leaving might require the presence of crews on ships, but crews could be shuttled between their ships and the docks.

      Or just use remote telemetry to do these operations. Ships already use harbor pilots who relieve the regular crew when in highly congested areas.

  • if they stopped putting so much time into their acronyms.
  • The more automation we put into aircraft, the more pilots are trained to use it. Consider instrument flight - a condition where the pilot is actively ignoring his or her limited senses and trusting data provided by sensing devices installed on the aircraft. Never mind what the pilot sees when looking out the window or feels from his sense of balance, when the instruments indicate they're passing a marker (outer, middle, inner), they make specific inputs to the aircraft's flight controls. When the instrum
  • by Jim Sadler (3430529) on Saturday April 19, 2014 @02:34PM (#46795969)
    The most promising method of creating advanced, military aircraft is getting human out of the craft completely. Human life requires, bulky, heavy, complex, life support considerations and limits the G forces that the craft can accept. Without people more weapons, more fuel, and more risk can be carried into battle without fer of human life being lost. We are at a tipping point at which any plane that has a pilot on board will fail to win in a conflict with an automated aircraft.
  • Will this system protect from the pilot pressing "gear-up" while the aircraft is still on the runway, and the plane collapses onto the tarmac, or is this tale merely apocryphal?

Real Users find the one combination of bizarre input values that shuts down the system for days.

Working...