Ecce: A tailless, stealthy UAV designed for carrier operations.
Pretty cool. It’ll be cooler still once they’ve demonstrated they can handle pitching deck.
|
|||||
Hot Mic
Bloggers in Arms
BlogrollContinuous WavePaid to BlogReciprocating EnginesSmarter'n MeWingmen
OmakaseAmazon Search |
Technology DemonstratorBy lex, on February 7th, 2011
42 comments to Technology Demonstrator |
Targets of Opportunityblog advertising is good for you Credo"Sign on, young man, and sail with me. The stature of our homeland is no more than the measure of ourselves. Our job is to keep her free. Our will is to keep the torch of freedom burning for all. To this solemn purpose we call on the young, the brave, the strong, and the free. Heed my call, Come to the sea. Come Sail with me." -- John Paul Jones "Pardon him, Theodotus; he is a barbarian, and thinks that the customs of his tribe and island are the laws of nature" --George Bernard Shaw, "Caesar and Cleopatra" "And those who were seen dancing were thought to be insane by those who could not hear the music."--Friedrich Nietzsche "A kind Providence has placed in our breasts a hatred of the unjust and cruel, in order that we may preserve ourselves from cruelty and injustice. They who bear cruelty, are accomplices in it. The pretended gentleness which excludes that charitable rancour, produces an indifference which is half an approbation. They never will love where they ought to love, who do not hate where they ought to hate."--Edmund Burke “You say that it is your custom to burn widows. Very well. We also have a custom: when men burn a woman alive, we tie a rope around their necks and we hang them. Build your funeral pyre; beside it, my carpenters will build a gallows. You may follow your custom. And then we will follow ours.”--General Sir Charles Napier "Μολὼν λαβέ" -- Leonidas "Blogito Ergo Sum" -- Neptunus Lex Amazon AssociateFor the Effort!Winnar! Subscribe CategoriesPagesTagsacademy
afghanistan
Araby
Blogging
buffoonery
culture
Defense Industry
economy
Flying
Friday Musings
geopol
Gratuitous slap
GWOT
Headlines
health care
History
iran
iraq
issues
libya
Media
Memory Lane
Military
Navy
norks
Oz
pakistan
piracy
politicians
politics
Politics and Culture
prc
pundits
science
sea stories
silliness
Small Stuff
SoCal
Tech Lust
technology
UAVs
UK
usaf
usmc
weapons
|
|||
|
Copyright © 2013 Neptunus Lex - All Rights Reserved |
|||||
Wow! What’s a reasonable estimate of how bad she was pitching? (Not having any experience in such matters)
A few questions from an uninitiated one: How will the unit costs compare to an F-18? Will it be able to maintain the same sortie rate?
This is a very perplexing development. Insurgents and their sympathizers hate existing U.S. drones. The few who have survived personalized ‘Hellfire’ attacks will soon be receiving better religious outreach on their way to martyr’s paradise.
Actually, two serious questions arise. Landing these new Grummans on a CVN is probably going to be a carrier-based performance.
1) Someone needs to start a pool – How long after deployment before the first one hits the drink?
2) Will stick jockies’ get 0, 1 or 2 freebies?
3) If no freebies, which of the following best applies?
a) No problem, GEICO will provide loss insurance for the “pilots”, if they have good driving records.
b) None of the stick jockies will get insurance, but under Obama HealthCare deductions to their pay will be required for the next 45 years.
c) The new craft will be flown by CIA types; if the operators screw up, they will just disappear. They never use their real names anyway.
Sounds more & more like we are living out the script of a movie…..(significant date adjusted)
The Terminator: The Skynet Funding Bill is passed. The system goes on-line August 4th, 2017. Human decisions are removed from strategic defense. Skynet begins to learn at a geometric rate. It becomes self-aware at 2:14 a.m. Eastern time, August 29th. In a panic, they try to pull the plug.
Sarah Connor: Skynet fights back.
The Terminator: Yes. It launches its missiles against the targets in Russia.
John Connor: Why attack Russia? Aren’t they our friends now?
The Terminator: Because Skynet knows the Russian counter-attack will eliminate its enemies over here.
John Connor: We’re not gonna make it, are we? People, I mean.
The Terminator: It’s in your nature to destroy yourselves.
John Connor: Yeah. Major drag, huh?
Much like the Forbin Project.
Question to all who have faced the challenge of pitching decks.
Why not a system on the carrier which provides the pitch angle time series at a rate of maybe 10 obs/sec, which is analyzed on-board the aircraft (or on-board the vessel) in real time to allow adjustment to approach speed. Touchdown could be synchronized with near-level deck using analysis of a time series amounting to a couple of minutes.
In these days of fly-by-wire where a computer must make high-speed adjustments to keep an inherently unstable airframe in the air — a system like that should be relatively simple to implement.
I have not seen a computerized autopilot system good enough to match human performance for a ship yet. With less than 2 hours on the helm of the USS Yorktown, I was doing better [smoother] than the “smartship” autopilot.
That was at only 15 knots stearring a CG in only 2 dimentions. You really think the systems are up to 10x the speed and operating over 3 dimentions? Remember that computers do not think, they do not anticipate, they react.
The problem with autopilots and computers is they are a reacting system and not capable of anticipating from the pattern they see. Add in the comunications delay to the reaction and …
Your fancy multi-billion dollar worth of development junk is more likely to fly the UAV into the stern of the ship than land it…
Actually, no. I suspect it will do tolerably well. Especially since the most advanced UAVs are not relying on a command link to tell them how to fly…merely what to do. The pilot is like a ship’s captain – he doesn’t have a control stick or rudder pedals.
So, how much money and caskets do we need to set aside for the UAVs that become enemy cruise missiles because the fancy computer system does not work as well as advertized?
Take off and flying – that is easy. Landing on a carrier is not. The tolerances are very fine.
Keep in mind that your fancy autolanding system has to be cheap – the only justification for UAVs vice manned is expendability. So no expensive system.
Sorry, I am not buying the advertizing. After all, if weapons worked as well as advertized – man would not have survived the bronze age…
“the only justification for UAVs vice manned is expendability.”
That’s not correct.
The primary justification for UAVs is that they will ultimately outperform manned aircraft, in some areas they already do. For instance how long they can stay out.
The fact that they can be lost without loss of life is a nice plus but it certainly isn’t the only justification.
Thunderdromes! Why risk the life of another Thunderbird?
The original Laserdisc featured a sampling rate of 44.056, or approximately 44 thousand samples/second in 1976. Super Audio CDs have as sampling rate of 2.8224Mz, or nearly 3 million times a second. Want to ask that question again?
True, you are speaking of random movements in three axes, but on the other hand, we have now have inexpensive CPUs which operate at over a billion cycles/second. Even if they “just react,” they have plenty of time to do so.
That’s not to mention the fact that there are several modern fighter designs which are dynamically unstable, and rely upon continuous fly-by-wire corrections in real time for stable flight, including the now “decrepit” F-117. Considering we have mature systems which can correct an unstable airframe in three dimensions in real time, one wonders whether adding a pitching-deck factor would create an unsustainable burden on the hardware/software?
…One recalls the story of an Israeli F-15 which suffered significant loss of control surfaces, but the “brain box” adapted to the catastrophic circumstances by dynamically altering control inputs to available control surfaces in order to maintain controlled flight.
So, yes, I think the hardware/software can handle the physics. It’s the strategic/tactical judgment of the aircraft AI which presents the greater problem. IMO.
OK, I’ll take a shot at the original question – those with experience newer than 1978 will probably correct me.
Changing aircraft speed: A/C on approach to a carrier fly at a constant airspeed. This optimizes the “angle of attack”, the angle of the aircraft wing relative to the air stream, at it’s most efficient. Adding power doesn’t change the speed through the air – it reduces the rate of descent, so that is done to correct a situation of being below the proper approach path. Reducing power increases rate of descent, and corrects for being above the approach path.
Keeping at the proper angle of attack (AOA) means the aircraft responds more effectively to changes in power. For carrier aircraft, the proper AOA also puts the tailhook at the proper level relative to the landing gear. Navy aircraft do not flare on landing – they just plow into the deck. If you try to flare you either miss the wire or catch the hook while you are still flying and really slam into the deck, often breaking something.
So, landing on a carrier at proper airspeed is critical in order to catch a wire. Getting fast or slow makes one more thing that has to be corrected. Getting too slow is really bad as it may not be possible to power out of that situation and climb back up to the glide slope before hitting the ramp. Could a computer correct it? Maybe – but it would probably be easier to move the glide slope, which is what happens now (or at least happened in 1970′s) in very heavy weather.
When the deck pitch exceeds what the optical landing system stabilizers can do, a device called the MOVLAS (Moveable Landing Aid System) was rigged. This presents the same light pattern as the regular Fresnel lens but it is controlled by the Landing Signal Officer (LSO).. make that a very very experienced LSO. The LSO judges where the aircraft is, how the deck is pitching, and moves the guideslope (as seen by the pilot) up or down such that when the plane is over the wires he’s at the same altitude as the deck. It’s very much an art to guide a plane down in these conditions. Any LSO’s out there care to comment?
In theory, a computer could estimate deck position based on a time series as you describe, plot drone position and speed based on radar, and adjust the glide slope for the drone accordingly. (I am presuming the ship beams a signal similar to the needles that show glideslope and centerline – which may be a false presumption). It might be easier than having the drone change airspeed from the optimal AOA on approach.
George V.
Edward: You are living an second order effect to a common theory to practice error for those entering Neptune’s realm for the first time
(i.e., baby boffins).
All waves are not sine waves. Ocean waves (and what they do to ships)are not one dimensional and on paper. They are a heterodyned result of multiple direction, multiple amplitude, varying oscillations of random origin, and unpredictable interactions at the air water interface.
Semi chaotic and very difficult to accurately measure or model in real time outside (usually several thousand miles outside) the laboratory. Hugely expensive to develop an algorithm to analyze and generate aircraft control surface adjustment signals in real time, and interact with wind gusts to correct a powered airframe trajectory.
Wetware can semi reliably bound the problem, but programming takes about 22-25 years per unit, and only a tiny percentage of available units are usable for a short service life. A few thousand out of a few billion at any given time. Those units are called Pilots.
God’s wind and God’s ocean eat aircraft and pilots, constantly.
Like major complicated, Dobie. Gonna cost a lotta bucks, might not work – except on calm days.
They’ll get it there. While it’s a tough problem, it’s still a physics problem and once they can nail down the appropriate inputs, it’s just a matter of feeding those inputs to an inbound machine that appropriately responds.
For a human the ship is moving fast, to a machine, it’s all very slow. The rate at which a machine can respond will make this doable.
Anyone who has doubts if the machines will be able to do the job should watch the videos of the cars that park themselves (parallel parking is one of the toughest parts of learning to drive) or ask the locals along the AF/PAK border what they think about the abilities of the “Beast of Kandahar” and his many drone friends…..I watched TOP GEAR last night on the BBC and they had a car that went once around the track and “learned” the track so it could drive around the track at over 100 mph by itself…..
The machines start out clunky but the engineers work out the bugs and they take off from there (no pun intended)
JR, you want to bet your life on that? Like Grampa says, the ocean’s waves are not all regular in nature; you can have three swells, especially in a crossing sea, that are almost the same and the fourth will be completely different. And it’s not always the fourth wave. Plus, what happens to the magic machine when we go to a total EMCON recovery? No auto-links, nothing. Oh, and GPS has been jammed or the satellites either soft or hard-killed?
I learned a very long time ago, the minute the engineers tell me they have an “elegant solution to a totally solvable problem,” to start applying Augustine’s Laws ASAP. When I see three combat deployments with no out-of-mean mishap rate losses, I will begin to relinquish my skepticism.
I can’t say about betting my life because I wont have the opportunity.
If the waves are irregular then what matters most is response time and there the machine always wins. Search for “robot dribbling” on youtube – I’m pretty sure I posted it on the flightdeck but that may be harder to find.
The other stuff you mention, I don’t know. I’m sure if we’ve thought about it chewing the fat here, then the engineers and such have thought of it too. I’m sure whatever solutions they may have happened upon are classified and that means I’ll never see it.
Think about all the arguments that can be made against operating a carrier at all. I did bet my life on the ability to recover very large fixed wing aircraft in all kinds of conditions, back when I was a young’in. More than one RAG squadron pilot gave me the sure impression that I was about to die.
Those are all problems that were overcome to an extent and I don’t doubt the same process will work here. Not tomorrow, but not thousands or even hundreds of years from now either. The idea that it is “too hard” for a computer to handle just doesn’t reflect the current state of the technology.
Comjam (and Grandpa), isn’t that what Fourier analysis is all about – to take an extremely complex wave and pick out the harmonics and their weights?
right then. I’m not ever contradicting anything you ever write again. I’m sticking with magic, myth, voodoo and legend.
Me thinks there’s enough mobile compute power and sensing technology (acceleration, positional, and even visual) available to tackle the problem of pitching deck.
Wetware cannot anticipate *random* events any better than software can, but software can react faster and more controllably than wetware given a properly modeled and functioning control system. Thus, the argument that SW is fundamentally disadvantaged because it can only “react” is bogus, IMHO.
Where SW is going to lose to wetware is in the ability to learn and adapt to conditions beyond the modeled control system. For example, a battle damaged engine or flight control is going to require a remarkably adaptive control system to work around such problems. This is where wetware shines the most and where the programmers will have to work the hardest, if they chose to handle such situations at all.
But, there are plenty of examples of autonomous systems learning the responsiveness of their control inputs and achieving stable operation in the presence of random external stimulus.
The problem Foobert, is control systems do “just react.” It is the nature of the beast when it comes to control systems. Anticipation requires intelligence, something we have been unable to supply to anything as yet.
Having said that, sometimes reaction is all you need. The pitching deck, however, is a fish of another color, however. Anyone that has handled a ship’s helm knows the problem one can have just maintaining a course. The JOOD during General Quarters (my DivO as it happens) needled me about steering a sinuous course in a heavy following sea. The fantail of that little DE was pitching as badly as any of the footage I’ve seen on You Tube of pitching deck carrier ops. Courtney didn’t have a flight deck, but Lester did, and I would not have wanted to be flying a Helo onto such a platform. Starch Wing would have been worse.
Frankly, I don’t see the problem being solved anytime soon.
Absent any other knowledge (i.e. being able to look out of the bridge and seeing the waves coming), anticipating a random event to happen and taking a preemptive action to counter that event *before* it actually does happen makes the action wrong almost all the time (unless you are suggesting some degree of clairvoyance on the part of the wetware).
I’m going to hazard a guess that at the helm control, there is visibility of the seas, offering a level of sensory input that an autopilot is not privy to. You may call this “anticipation”, but, I’d call it superior sensory input. Wetware’s ability to process visual cues is an amazingly powerful tool that we may never be able to fully replicate.
I’m gonna agree with Foobert here about the main issue being the amount of sensory data. If we can create a machine that gathers data from enough sources (including visual) we can easily solve the problem. The other part of it is gathering data over a period of time. Sure, a drone that only looks at the current position of the deck is going to be a lot less accurate than a pilot who has a general sense of how the deck is moving and has past experience to go on. On the other hand we can easily design a system that takes that into account. Heck, it’d be possible to put sensors on the ship that create a model of the deck’s movement over time and communicate directly with the drone. At that point it could literally talk itself down. Much less room for error if implemented correctly.
There are only two reasons I can think of to take a pilot out of the covkpit.
1.) Where there is no realistic expectation of the aircraft returning from it’s assigned mission.
2.) Where mission requirements mandate a vehicle capable of maneuvering with G-forced which would be unsurvivable by a human aboard.
My main concern with unmanned operations is that we become too adventurous in our use of force, since we no longer are risking our own people’s lives.
When we have organic skin in the game, then risk management is a serious issue. When there is no organic skin, then it becomes a risk of purely financial and material resources.
There’s a third reason. Extremely long missions. Once you get to multiple shifts, unmanned (more correctly, remotely manned) becomes very useful. Especially when you are operating at extreme altitudes.
Or you could haul them out and buy them a pint of Guinness.
Your call.
The best flyers I’ve ever seen have bird-brains. Can’t help it, they’re birds. So we just need a computer as smart as a swallow.
Do you mean an African or a European swallow?
On problems encountered on/off flight decks: http://www.youtube.com/watch?v=If9dJs7W3qI&NR=1
laden or unladen?
Thanks, fellows. All had very good points and it was the discussion that I hoped to spark among those who have faced the beastly tricks of Mother Nature and Father Finagle.
I don’t think the pitching deck problem will be easy, but I suspect it’s going to be solvable. Yes, we’ll lose a few expensive drones along the way, but the fact of the matter is we lose aircraft in tough conditions already. In my opinion the the success or failure of the implementation is going to be largely administrative in nature. If Navair has a moment similar to the Air Force being told they would get on board with UAVs (remember some generals lost their jobs over foot dragging on that) the progress will be more rapid. As a helicopter pilot I sympathize with the Naval Aviators concerns here, I would have never believed that an effect UAV could be built off an R-22 until I saw it done. The only question is what kind of missions are we looking at? Recon, obviously, that’s been going on since Viet Nam at least. Predator and Reaper have demonstrated that the attack and CAS missions are within reach. Fleet defense? I don’t know how long it would take to develop air to air capability, but it might not be as long as we would expect. Interesting times…
Despite all the sage comments about it landing aboard an aircraft carrier – is it not supposed to be a “war bird” rather than a “carrier-qual” bird?
While I am unfamiliar with all its capabilities, if I were an enemy fighter pilot, I think it would be a Marianas turkey shoot for me against these things if they ever flew over my homeland, regardless of any stealth technology and other ‘pmf’ stuff.
But if I bagged even 25 of these expensive dumb suckers, I doubt the commissar would call me an “Ace”. Would they? Just say’n.
Well the X-47B is supposed to be a tech demo platform. Something to do the, “Hey, now that we have proven this thing can land, taxi, and take off with only minimual input from the operator back in Fallon. Lets try to have it park upload an pair of AMRAAMs and Sidewinders, then do the previous launch, taxi, recovery.” However, there are already some folks in both NavAir but also the USMC beating the farm on the X-47B (or its grandchildren) to be the follow on to the (or augement) too the EA-18G. Sitting through a brief in the EW world, using all sorts of buzzword bingo, I heard some talk about an F-35 drone controller that was in control of 4-5 UCAV’s with ARM’s and EW systems. The UCAVs would open the holes and take shots at pop-ups, while the EF-35 would be the master controller and direct specific attacks as needed while the strike package goes in.
What really needs to happen though we need someone like Captain Jim Reeves to come in and start to ask the serious questions about what we can and can not do. Then turn around and start to answer the questions at the same time in real field conditions. A repeat of the various fleet exercises using drones, like we did in the 1920′s and 1930′s with the carriers will start us on the path of deciding how to use these right. All I have been hearing or reading in magazines like The Hook, Proceedings, JFQ, and others has been “Drones are the future and will be able to do X,Y,Z, and the Kitchen Sink….” Yet there hasn’t been any proof beyond some spotty usage here and there in military service.
Don’t get me started on the overselling of unmanned aviation (mostly by people unfamiliar with the field). I’ve been mulling over a Proceedings article on the subject.
“The Facts of Life About UAVs”.
Now, you CAN have something like UCLASS motoring around shooting HARMs – and if it takes a hit, the operator will be most unhappy…but not nearly as unhappy as if he was a POW or worse.
Mark my word, the JSF (if it survives the budget cuts) will be the last manned fighter bought.
By time the JSF or it’s budgetary replacement enters it’s midterm life span, technology will overcome the man in the cockpit. It is already happend in my life time and career.
Redeye: The cost of modeling “pitching deck” will be part of the budget getting cut. Lotta code to be written and then tested in the real ocean. The chart is not the ocean and the model is not the phenomena. This won’t be done on the cheap.
I think “need input” is at the theoretical design level now, and will be for quite some time to come. I think bird brains are a lot more sophisticated than we give them credit for.
If it was as easy as you say, DASH would have been a world beater, instead of just a suicidal egg beater.
Remember I am talking 20-25 years out. Administrations and technology change. I hope to be alive to see it but I believe in my heart the JSF is the last manned fighter we buy.
The code has been written. The pitching deck is caused by the swell and the wave and these are easily computable and subject to near exact forecast. It’s like a fire control system that exactly computes the movement of the target in 3 dimensions. What has been doing the orbital rendezvous for decades? What lands the space shuttle?
It’s an interesting bit of simulator testing.
Code written? Don’t doubt a lot of it has and works pretty good in routine conditions. I thought we were talking about not so normal…gusty winds, confused seas, high waves, complex swell and just plain damn ROUGH.
What’s the code to give you power, power!, POWER! and avoid RAMP STRIKE and fantail splat/explode/burn? Remember 99% success rate means no drones left half way through time on station for a fairly vigorous campaign, at least through the first few deployments.
I’m from Missouri on time and cash burn through rate to a proven, reliable, trusted system. This path through the Primroses is well worn.
GB,
This is where I toss off an “it’ll be easy” and move along smartly. I sure can’t do it but I’m pretty sure that smart guys can/have. The system would be sampling the environmental factors at a phenomenal rate and have the range of motion in all 3 axis down to the nanometer. (unless SPAWAR gets involved in which case it would be down to the mile and doomed.)
X-47B UCAS First Flight Music Video: http://www.youtube.com/watch?feature=player_embedded&v=UIq5dT7D_ic
“Two minute musical revue of historic first flight of the U.S. Navy’s X-47B Unmanned Combat Air System demonstration aircraft, designed and built by Northrop Grumman Corporation. Flight occurred Feb. 4, 2011 at Edwards AFB, Calif.”