back to article US Navy robot stealth fighter in first unmanned carrier landings

The last of the great barriers has fallen: an unmanned aircraft has successfully made autonomous arrested landings on the deck of an aircraft carrier. It's now plain that robots are not just as good as human pilots - they are as good as the best human pilots. As most regular Reg readers will be aware, the X-47B project …

COMMENTS

This topic is closed for new posts.

Page:

    1. Peter Ford

      I always assumed that was more a signal to the catapult operator that everything was ready to go, and warning the pilot that she's about to be launched...

      In this case the robot probably gets some other signal, unless the software is designed to assume that the shove in the back means "let's go!"

      1. Allonymous Coward
        IT Angle

        the robot probably gets some other signal

        I wonder if he could use a mouse with his other hand?

  1. Tom_

    Now they should work on getting rid of the meat from the carrier as well.

  2. John Smith 19 Gold badge
    Terminator

    So "Tinman" is a go.

    Would could possibly go wrong?

    1. AceRimmer1980
      Pirate

      Re: So "Tinman" is a go.

      It will prove itself undefeatable in air combat, but will be busted by the RIAA, when it downloads every MP3

  3. MisterIrrelevant

    What about AEW?

    Surely a potentially even greater use of unmanned carrier aircraft would be a replacement for the E2? With no crew to rotate, more fuel capacity and the possibility of unmanned aerial refuelling a carried could potentially keep its radar coverage on station for days at a time. Indeed - it could be manned around the clock from land bases and free up a few bunks...

    1. bep

      Re: What about AEW?

      Quite, all the things that are 'routine' but still require carrier qualified pilots, like COD (Carrier on Board Delivery, at least that's what they used to call it), in other words the mail and beer run (and high value spares, of course). Crying out for a suitable drone, and the AEW thing has got to be the next option shirley?

      The broader point is that, once you no longer have overwhelming superiority, dispersal and muliplying your launch platforms becomes a lot more sensible. Smaller carriers carrying lots of drones give the potential enemy a different kind of headache. This is because the aircraft, whether manned or not, are still relatively fragile and if your big carrier suffers heavy damage you're likely to lose the lot.

  4. Jerry H. Appel
    Terminator

    Still not fully baked, but interesting.

    I agree with others that the test was successful in benign conditions. Others have mentioned the dangers of ejection, and as I have a friend who is a retired Marine Corp aviator and was night-landing qualified, they are all true. One ejection and it is all over for your back, and his just after a launch.

    Now as to the visual signal by the deck crew, according to Aviation Week & Space Technology, the deck controllers use an arm-mounted device to signal the robot what to do and where to go once on deck, nothing about launch, but I imagine someone has to signal the robot to go from idle to full-throttle. Don't you? Pilots drive these birds on the flight deck, so the robot has to respond to such signals.

    Agree that the UCLASS will be the Navy's answer for a stealthy strike/interdiction machine. F-18s do not cut the mustard in that arena, at all. Look at all the stuff hanging off the wings and fuselage! Really!

    I also agree that a UCLASS can certainly outmaneuver a man-driven aircraft given that it can be made both stealthy and extremely agile. You see the problem with stealthy flying wings is that they don't handle high alphas well or rapid directional changes well. If they did then they wouldn't be stealthy. Ever see a B-2 perform an aileron roll, a loop? Remember that Boeing 367 was rolled on an early flight to prove it was good enough, but no one is doing that with one of these stealthy robots . . . yet. Think about such a wing-based aircraft dealing with a dogfight situation with multiple bogies and missiles flying all over the place. Just try programming that!

    1. Marcelo Rodrigues
      Devil

      Re: Still not fully baked, but interesting.

      "I also agree that a UCLASS can certainly outmaneuver a man-driven aircraft given that it can be made both stealthy and extremely agile. You see the problem with stealthy flying wings is that they don't handle high alphas well or rapid directional changes well. If they did then they wouldn't be stealthy. Ever see a B-2 perform an aileron roll, a loop? Remember that Boeing 367 was rolled on an early flight to prove it was good enough, but no one is doing that with one of these stealthy robots . . . yet. Think about such a wing-based aircraft dealing with a dogfight situation with multiple bogies and missiles flying all over the place. Just try programming that!"

      I tought that F-22s and F-35s were quite agile. Aren't they? The B-2 is, indeed, more a flying beetle than anything else - but it is outdated technology. Their newer brothers are much better.

      About the dogfight... in my opinion they would be MUCH better than humans. Not today, not tomorrow. But give it a few years of research and development...

      A computer doesn't get nervous. It doesn't get distracted, and it have no problem prioritizing threats. It can, even, decide that self-sacrifice would be a good idea in a no-winning scenario. All it would have to do is avoid enemy fire long enough to do a kamikaze on target.

      Like this: (in pseudo code)

      "enemy fire detected. -> 5 missiles incomming. -> 3 enemy fighters and an enemy bomber (my target). -> evasion deemed impossible, no escape solution. -> alternate solution: evade long enough to explode/hit the target. -> EXECUTE."

      The bigger problem isn't the capability/feasibility - it is just that people are afraid of let the control go.

      1. Jerry H. Appel

        Re: Still not fully baked, but interesting.

        Oh, you make the programming seem so simple when all the actors are moving in all four dimensions and the UCLASS is acting alone, but I do agree about the robot being able to unblinkingly sacrifice itself. Of course given a quantum CPU this all takes no time, but we are not there yet. As to doing the kamikaze thing, remember there will be overlords/operators watching who just might override the robots decision making.

  5. b0hem1us

    What is so new about this?

    Autopilots have been landing on carriers forever. As a matter of fact our dynamics and controls professor at the university told us they had to program in an error because the autopilot was so good that the catching hook made a huge groove on the deck from landing on the same precise spot all the damn time.

    1. Alan Brown Silver badge

      Re: What is so new about this?

      This raises the questions:

      Is the autopoiot trusted enough to do this in shitty weather?

      What is the failure/go-round rate compared with flying in on fully manual control?

      Fuzzing the landing is also applied to civilian ILS autoland systems for the same reason - it was tearing up the same strip of runway.

    2. SkippyBing

      Re: What is so new about this?

      Indeed, I believe it's actually a version of the auto land used in the Super Hornet.

      The problem is, while you have human pilots, they have to keep in practice for those occasions when the auto-land doesn't work* which means most landings are done manually anyway...

      *Recent events in San Francisco indicate the civilian world may want to re-look at this.

  6. Rusty 1
    Happy

    Now what?

    Can't help thinking at the end of the first video of the drone landing that everyone is standing around waiting for the pilot to raise the canopy or do something interesting.

    Could be there a while...

  7. Anonymous Coward
    Anonymous Coward

    Oh I've been waiting for this moment

    The next time I'm up against some flyboy trying to cock-block me with tales of carrier landings. i'm gonna say ' Big deal, my phone can do that now'

  8. Tom 11

    One step closer to forever,

    Haldeman eat your heart out!

  9. SkippyBing

    Thought for the Day

    Is this surprising, not really, Carrier Auto-land is nothing new, F-4 Phantoms had it in the '60s.

    However, UAVs are currently flavour of the month with the politicians because it reduces the likelihood of casualties to your own side because they're not in the plane, at the same time they require large amounts of bandwidth to transmit all the data back to wherever to allow someone to make a decision.

    The answer to the latter problem is autonomous UAVs which do all the decision making themselves. The question is, will any politician outside of an all out war let that happen? Certainly in the Vietnam war when the US forces had the capability to destroy enemy aircraft before they could be seen this was deemed unacceptable and you had to have visual confirmation of the target's identity before firing. So despite having spent $Billions developing a capability and doctrine to kill the enemy at range the pilots were forced to close to a position where they were at a disadvantage before they could open fire. You saw a similar thing with the Osama operation where Barrack Obama had a real time video feed of the operation, giving him the ultimate long screwdriver to interfere with the mission.

    Ultimately I can't see politicians accepting a drone making decisions about targeting, especially if there's no way to interfere. The problem is if they've already replaced all the piloted aircraft with drones to save money that could leave them defenceless.

  10. briesmith

    Carrier Nonsense

    All of this ignores the decreasing, rapidly decreasing, survivability of aircraft carriers. Within a procurement generation - 10 to 15 years - no carrier will have a combat life expectancy sufficient to justify its exposure to enemy contact/interdiction. What the Japanese did to the Prince of Wales and Repulse in 15 minutes in 1942 the Somalis will be able to do the Nimitz, Reagan etc in 2030 and God help the poor Queen Elizabeth.(I won't mention the name of her sister ship out of sensitivity to people's feelings) which will be sailing with none of the protection the US can give to its carriers.

    The same could be said to apply to fast jets. Their development, build and deployment costs continue to outstretch the costs of killing them to the extent that purposeful affordability has almost disappeared.

    So, how do we apply our technology in a way that is effective, durable and affordable?

    We build or take up from trade, cheap general purpose ships, harden them a bit ("militarise" them) and use helicopters to launch and recover UAVs. As a refinement of launching we could use catapults in a similar way CAM merchantmen launched single use Hurricanes in WW2 leaving just recovery to helicopter capture.

    In this way we can have lots of drones on lots of ships doing lots of different things all very cheaply and in an eminently scalable fashion in the event of conflict.

    (Some fleeting Googling reveals a lot of weapon systems (examples I found include the Bushmaster and Goalkeeper systems) are already containerised and suited to deployment on temporary "warships",)

    This is an affordable war fighting future where brains, imagination and adaptability (I know, not noted British qualities) and not money will rule.

    1. Matt Bryant Silver badge
      FAIL

      Re: briesmith Re: Carrier Nonsense

      You need to go read up on Duncan Sandys, he was wrong and so are you. Every time thee is some new development that makes combat aircraft or carriers "obsolete", a countermeasure is found. And if you want British ingenuity, go read up on the British Sea Dart missile system that the Royal Navy used to save the USS Missouri's bacon, after the USN's own countermeasures failed.

    2. SkippyBing

      Re: Carrier Nonsense

      So your proposal is:

      1. Stop building military vessels with integrated defences and networked escort vessels because they're too vulnerable.

      2. Replace them with converted merchant shipping which for some reason isn't, despite not having any defences.

      3. Add some containerised defences, much like the ones on the naval ships you got rid on in 1, in fact the two examples you listed are already on Royal Navy ships.

      Oddly this to my mind puts you back to square one, just without the survivability of a naval vessel in the event of actually being hit.

      There may well be an imminent revolution in naval affairs, this isn't it.

  11. briesmith

    Carrier Nonsense - The Sequel

    I know, it was 1941. Couldn't find an Edit button. Sorry for the carelessness. (It also took a little longer than 15 minutes to polish off the PoW - a totally new ship commissioned only a few years earlier and already the victim of a seeing-to by the Bismarck) - and her much older consort, Repulse but I was trying to make a point about vulnerability.)

Page:

This topic is closed for new posts.

Other stories you might like