Flying_727 From United States of America, joined Jun 1999, 428 posts, RR: 5 Posted (13 years 3 weeks 5 days 7 hours ago) and read 1745 times:
There are no pilots flying the plane? Today on the Discovery Channel they talked about the 777 and how some day there will only be one just checking the computer. I for one hope it never happens but how long do you think it will take.
Aaron atp From United States of America, joined Mar 2000, 533 posts, RR: 2 Reply 1, posted (13 years 3 weeks 5 days 7 hours ago) and read 1608 times:
I made a post about this when someone proposed the BS about restoring the Flight Engineer to the cockpit and having 3 man cockpits because they think it's safer that way.
I think we'll have 1 man cockpits when the bugs get worked out of WAAS and LAAS for gps. Then we'll start flying the direct routes everyone is talking about. I'm guessing that all new aircraft built after 2015 will be one man cockpits. Give it another 25-50 years after that for an unsupervised automated cockpit, but maybe by only 2025 we'll have supervised, but fully automated cockpits.
AC_A340 From Canada, joined Sep 1999, 2251 posts, RR: 2 Reply 2, posted (13 years 3 weeks 5 days 5 hours ago) and read 1590 times:
After reading this topic, I checked my email where I found this joke. It fits quite nicely.
The World's First Fully Computerised Airliner
> The world's first fully computerised airliner was
> ready for its maiden flight with out pilots or crew.
> The plane taxied to the loading area automatically,
> its doors opened automatically, the steps came
> out automatically. The passengers boarded the plane
> and took their seats.
> The steps retreated automatically, the doors closed,
> and the airplane taxied toward the runway.
> "Good afternoon, ladies and gentleman," a voice
> intoned as the airplane lifted off. "Welcome to the
> debut of the world's first fully computerised
> airliner. Everything on this aircraft is run
> Just sit back and relax. Nothing can go
> wrong........nothing can go wrong......nothing can
> go wrong......"
FDXmech From United States of America, joined Mar 2000, 3251 posts, RR: 38 Reply 3, posted (13 years 3 weeks 5 days 5 hours ago) and read 1583 times:
It strikes me funny how people (especially pilots such as on the 777 show) can talk about unmanned cockpits even before there is even mention of unmanned busses or even taxicabs. I think this type of talk is extremely detrimental to the aviation profession as a whole and totally without merit. When flight crews such as on that show talk like that, it appears to me they are trying to act in a self depreciating manner in a backhanded sort of way. I think its going to bite future pilots on the **s during contract time and diminish the extreme importance of the job they perform today.
D L X From United States of America, joined May 1999, 10561 posts, RR: 53 Reply 4, posted (13 years 3 weeks 5 days 4 hours ago) and read 1565 times:
Re: unmanned pilots before road vehicles.
Well, the fact is, it is easier to navigate through air than on the ground. On the ground, you have to deal with roads. That's a very tight bound that vehicles have to stick to. (The average road is only 50% wider than the vehicles that travel on it.) Planes are quite a bit more forgiving (except when landing or taking off, but we've already conquered the auto-land facility).
Also note that there are an awful lot of rockets and other space travelling vehicles that are unmanned. In fact, the unmanned ones came first.
FDXmech From United States of America, joined Mar 2000, 3251 posts, RR: 38 Reply 5, posted (13 years 3 weeks 5 days 3 hours ago) and read 1554 times:
So your telling me its easier to navigate roads than through the air. Thats a new one on me. I guess the engineering on cars is a lot more sophisticated then I thought. Conquered autoland? I don't think so. Yes, the new aircraft have very reliable autopilots but 100% reliable - no way, I see them malfunction all the time. Ever hear the term "unsat cat 3 landing" which means the pilot has to intervene during autoland, I have and its not uncommon (though less frequent than in the past). Guess what, who's going to land the plane now in our unmanned cockpit. What happens when you have other system problems which are fairly common. Remember this isn't a moon mission where there is a room full of people concentrating on 1 rocket. And speaking of unmanned rockets, thats the point, they are unmanned which means no one is onboard. Do you think an astronaut will fly in a rocket without any means of controlling it if required,I think not. Remember flying people in an airline environment is not an experimental excercise, the flying public deserves better then that.
D L X From United States of America, joined May 1999, 10561 posts, RR: 53 Reply 6, posted (13 years 3 weeks 5 days 3 hours ago) and read 1545 times:
You've really lost me. I'm not telling you that it is easier to navigate roads than through the air. I'm saying the exact opposite. Yes, I know that while autopilots are good, they are not perfect. YET. No one is saying that tomorrow there will be pilotless aircraft, but it is a possibility soon, as in 10-15 years from now.
And don't think for a second that those unmanned space missions are pointless. There are a lot of people that are very concerned that the payload on the rocket gets to where it is supposed to.
Besides that, if airlines go pilotless, there WILL be a whole room of people dealing with the many flights in the air, or there will be a whole room of computers dealing with the many flights in the air. (Unless they opt for embodied intelligence, as opposed to centralized control. E.I. in my opinion would be the much better way to go. I'll explain that one in another post if you want.)
Do I think an astronaut will fly in a rocket with no means of controlling it? Yes. But that's not a good analogy anyway. Would a paying passenger fly in a plane with no way of controlling it? Well, they have no way of controlling it today...
FDXmech From United States of America, joined Mar 2000, 3251 posts, RR: 38 Reply 7, posted (13 years 3 weeks 5 days 2 hours ago) and read 1539 times:
That was a mistake on my part. I'm incredulous you think its easier to navigate in the air than on the roads. Is an instrument rating required to drive your car on a foggy night to the grocery store? Haha. Each time a plane flies their is someone directly accountable for the safety of that flight and that person is the captain. Who will be directly accountable for the safety of flight in the unmanned cockpit. A programmer on the ground or some other nameless company official? And whats to gain. Also, I dont understand your comparison of the passenger being unable to control the airplane, thats the pilots job. Do you feel the average pax would rather fly the plane himself? Even freight trains traveling 20mph aren't unmanned and I don't hear any future ideas to the contrary. Also I didn't say or mean to imply unmanned space flight is pointless. But speaking of that, how many unmanned rockets have crashed in the past several years to the point that they are barely insurable. Theres too many variables on these airplanes to go wrong and they do, all the time. Thats why you need the primadonnas up front.
D L X From United States of America, joined May 1999, 10561 posts, RR: 53 Reply 8, posted (13 years 3 weeks 4 days 14 hours ago) and read 1517 times:
"I'm incredulous you think its easier to navigate in the air than on the roads."
Come now. Why do we have autopilots for planes, but not for roads? Because figuring out where the lines on the road are is an incredibly simple task for a human, but an incredibly hard task for a computer. An autopilot can fly any time of day, in pretty much any weather, or at least well beyond the point at which a person can see. A computer pretty much only operates via instruments. These tools are highly refined in airliners, but there is no such equivalent in a car. That is why a computer could fly a plane a lot easier than a computer could drive a car.
What's to gain by removing the pilots from the plane? You must know how much pilots make. Not having to pay for pilots would be an incredible cost savings for the airline.
You don't understand my point about the pax not flying the plane, because it is the pilots job. The point was that you suggested no pilot would want to be on a plane that he could not control. Well, that's not a problem. There just won't be any pilot on the plane then.
Freight trains are manned, correct. Why? There are things that the computers can't see so easily. (As you may be able to tell, we have not been able to build a computer that can see well at all.) What if a car is crossing a track up ahead, and the train might hit it. Or, what if it's just a person walking across the track? There are no people to walk across the flight paths at 35000 feet. For a better analogy, note that many subway systems *are* unmanned. There's simply no place for people to get in the way there.
Lastly you mention the number of rockets that have crashed. I think this number really pales in comparison to how many rockets have safely delivered their payload to space. Think about just how many satellites (unmanned vehicles!) are floating around in space adjusting their trajectories as need. Note how long they've been up there without problems.
Matt D From United States of America, joined Nov 1999, 9502 posts, RR: 51 Reply 9, posted (13 years 3 weeks 4 days 13 hours ago) and read 1509 times:
Why no mention of the real reason the airlines and mfrs are striving to have fully automated flight decks? COST!!!! I'm convinced that the real agenda is not to enhance safety, but to be able to eliminate the pilots altogether. Don't most pilots make 6 digit salaries? Think of how much extra money the shareholders will be able to pocket if suddenly they didn't have to pay salaries for thousands of pilots anymore. The initial cost of purchasing automated flight desks will pay for itself in a very, very short time. The rest will be "gravy"....
FDXmech From United States of America, joined Mar 2000, 3251 posts, RR: 38 Reply 10, posted (13 years 3 weeks 4 days 12 hours ago) and read 1505 times:
Autopilots are incredibly complex, expensive engineering marvels developed over many years to the point they are at now. To me the development of a hypothetical self guided car would be relatively simple in comparison. Put imbedded sensors in the road. The challenges to an unmanned train would appear even simpler to overcome. My big argument is that you are taking one of the most complex machines built operating in one of the most complex operating environments and telling me it is the easiest of transportation modes to remove the human element. I suggest you become a bit more technically and operationally fluent before making these rash assumptions. By the way DLX, my postings to you are in no way meant to be derogatory. I always enjoy reading your threads.
Aaron atp From United States of America, joined Mar 2000, 533 posts, RR: 2 Reply 11, posted (13 years 3 weeks 4 days 1 hour ago) and read 1476 times:
I could write my doctoral dissertation on the advances that would be needed in navigation, engineering, quality control, and human faith to implement an unsupervised fully automated aircraft control system.
I am not.
However, I do believe that between 2050 and 2065 we will be installing fully automated, unsupervised cockpits.
Do you really think that you will hear "unsat cat 3 landing" in 50 years? No. We will advance; we are advancing; the rate of advancement is increasing.
>>>this type of talk is extremely detrimental to the aviation profession as a whole and totally without merit. When flight crews such as on that show talk like that, it appears to me they are trying to act in a self depreciating manner in a backhanded sort of way.
Totally without merit? self depreciating? I take that as a personal attack. I do not understand how visions of positive advancements in the industry can be detrimental unless you are an anti-technology radical who feels that computers will one day take over the world and kill all carbon-based life forms.
Do you also think "predictions" by Jules Verne, Isaac Asimov, and Arthur Clarke had a detrimental effect on our society? Would Woerner von Braun have masterminded the first German rocket or the Saturn-V without reading science fiction when he was young; which predicted men on the moon before rockets were ever created. Would we be pilots today if our predecessors had not dreamed of future innovations?
>>>I think its going to bite future pilots on the **s during contract time and diminish the extreme importance of the job they perform today
If job security for your children is what you are worried about, take a look at Detroit. Machines doing a job formerly done by men. Should we attempt to preserve a year 2000 job market by sabotaging visions of the future? Maybe Flying_727 is a future engineer who may one day design these automation systems; and maybe our predictions will influence his dreams of the future. Without a goal, there is no advancement.
D L X From United States of America, joined May 1999, 10561 posts, RR: 53 Reply 13, posted (13 years 3 weeks 4 days ago) and read 1459 times:
FDXmech, I'm honored that you feel my posts are interesting, and I take no offense at your statements. Except this one: I suggest you become a bit more technically and operationally fluent before making these rash assumptions.
Ouch. Well, you may know, that I am an engineer. I would consider myself pretty technically fluent.
How do planes get from A to B today? (After takeoff, and before landing.) With the aid of an autopilot. Think about a long flight to somewhere where there aren't many other planes. I'm talking South America to Australia. No real nasty ATC to deal with, so the pilot puts the plane in the air, lets the nav take it up and out near the south pole, then back north towards Oz. Nearing Sydney, the pilot listens in on the radio and receives directions from the controller so he fits into the landing pattern, and ultimately is set up to land the jet.
So, the cruise is all handled by the autopilot, right? So, that part's out of the way. Now, how about that insertion into the pattern. Is there any real reason why that contoller has to be human? Not really. make it a computer that looks at where all the planes are, and then tells each plane where to go. It sends some electronic signal that gets picked up by the plane's autopilot, and the autopilot adjusts the attitude and points the plane on its new trajectory. The plane gets close in, lines up, and autolands.
From what I understand, not being a pilot, there is currently no such thing as auto-takeoff, just autoland. (Because takeoffs are actually more challenging than landings? Aaron?) However, I can't imagine our not being able to tackle that problem given 10-15 years. And, then we have fully automated flights from takeoff to landing.
Would I fly on one? Hmm... Now, that's another story. I wouldn't be the first. The type of engineering I do is computer engineering, specifically artificial intelligence, at least today. Computers aren't perfect, and weird things happen sometimes when you thought you had something bug free...
Now, back to the autocar and autoroads for a second. Putting beacons (I assume that's what you meant) in the road would certainly make the problem much easier. However, that is an impossibility. Maybe possible on the major highways and biways that connect cities out on open road, but not on all roads. The cost of putting the beacons in the road is one thing. The cost of maintaining them is... well, I wouldn't want to pay it. There are many places in this country where they can't even put reflectors in the road because of the maintenance cost. Then, on top of that, the beacons won't tell you where the other traffic is. If the road is straight (which they aren't in Joisey ) the problem gets pretty easy with radar, but once the roads get curvy or hilly, range finding gets near impossible. Use transponders so the cars can talk to each other? That would work, but it won't help you when you're about to hit a deer.
Prebennorholm From Denmark, joined Mar 2000, 6019 posts, RR: 55 Reply 14, posted (13 years 3 weeks 3 days 22 hours ago) and read 1448 times:
For many, many years we have been working on an unmanned metro here in Copenhagen. It doesn't work. When the metro begins operation in 2003 the trains will be manned.
There is still a long way to go to year 1984. Jules Verne was born too early in a world too antique.
The problem is not automation. The problem is failure management in systems which must not fail.
The comparison to space vehicles is not relevant. A lot of people and resources would be saved if one clever man in the spacecraft could do the job which is performed by a hundred clever men on the ground. The problem is that sometimes it is not very practical to put a man in a spacecraft. Therefore he is substituted by a hundred men on the ground.
Rome was not built in one day. Fully automated passenger planes will not be even considered in our lifetime. The most important reason for that is that there will be too few passengers willing to board such planes. Therefore nobody will risk their money on developing such planes.
There are limited economic resources for such developments. The resources will go where the the investments will pay back. Automated passenger planes won't pay back.
Always keep your number of landings equal to your number of take-offs, Preben Norholm
Aaron atp From United States of America, joined Mar 2000, 533 posts, RR: 2 Reply 15, posted (13 years 3 weeks 3 days 15 hours ago) and read 1419 times:
today the worlds fastest computer processes at a rate of 1.3 trillion operations per second. We cannot define how many operations per second the brain is capable of, because the brain is not a processor, it is a conscious being. Today, a conscious being in control of the aircraft is more "useful" than a computer control system based on current technology. Read the post by Prebennorholm again if you doubt that fact. But were are trying to look 50 years into the future.
The current trend of advanced (commercial and military) aircraft is fly-by-wire control systems. Why? Because these aircraft have such complex handling that simple human manipulation of the control surfaces is inadequate. Perhaps a human could fly the same plane with mechanical controls, but they would be too preoccupied to accomplish any other duties. Even in straight and level flight.
Mechanically controlled aircraft are becoming outdated. Perhaps they are not outdated today, but please remember that we are looking 50-75 years in the future. When WAAS and LAAS are implemented over the next few years, it will only be a short amount of time before every airport in the world has a precision GPS approach. When this is fully implemented, approaches will be vectors defined in finite coordinate space, accurate to within less than a meter. In the not-to-distant future, approach plates will no longer be needed, their equivalents will be electronic chart displays.
So I'll return to my original estimate, that by 2025 we will have fully automated, but supervised, cockpits. 25 years after that, I think the cockpit will be useless. Aircraft will be so complex, that if a failure were to occur that would render the "computers" uncapable of landing the aircraft, no human alive would be able to do anything but observe the travesty. This means 10^N redundant systems have failed, and the aircraft has found no method to control the aircraft using available systems.
Let's take a look at United flight 232 in 1989. At FL370 the Number 2 fan on the DC-10 disentigrated, leaving the aircraft without functioning hydraulic control. first, by talking about this, I want to say Capt Haynes and his crew made one of the most heroic acts ever made by any flight crew. This crew and a DC-10 instructor who was a pax landed the aircraft using only engine thrust to control the aircraft. Not to reduce value of their effort, but I think that the aircraft control systems of the future will be able to do the same thing. Aircraft systems would learn. Redundancy isn't enough to guarantee safety. What we will have is aircraft systems that have the capability to learn the handling characteristics of the aircraft. If a major system fails, it doesn't use the same set of commands to execute a maneuver. Instead it adapts to the situation. Intelligence is what you need to make a better autopilot. Maybe it isn't "smarter" than humans, but it is more capable of analyzing the situation, reacting quicker, and adapting to the situation. At any given time, it will "know" the entire operatiing envelope of the aircraft. It certainly won't be AI as it has been presented in science fiction, but it will be adaptive and flexible.
I hope I haven't lost my original point here, but try to imagine an aircraft design that exceeds the capability of a human pilot. A pilot in the plane would be useless.
It is irrelevant to compare automobiles, trains, or any other method of transportation to our debate, because none of these vehicles have reached the level of technical advancement that has been implemented in aircraft.
D L X From United States of America, joined May 1999, 10561 posts, RR: 53 Reply 16, posted (13 years 3 weeks 3 days 14 hours ago) and read 1414 times:
While I agree that road travel and rail travel are not good analogies to how a plane could be automated, I think people may fail to see why space travel is. Unfortunately, the perception of space travel is that there is a room full of 100 or so people for each flight into space. That's not quite the case (anymore) and it also shows that we are limiting our thought to rockets. There is a lot more than rockets in space travel. Remember the Mars rover (the one that actually worked...)? It was unmanned. The distance from Mars to Earth is such that it would take a long time for any signal from earth to get to Mars and back, so controlling the rover from home would be very inefficient, if not impossible. So, what did Prof. Rodney Brooks of MIT do to solve this problem? The rover was completely autonomous. It was able to tell what should be done at any moment, and the 20 or so people in a room at JPL gave it basic instructions like "go check that thing over there out" and "can we reexamine that rock 200m west?" The onboard computer figured out exactly how to get there, wether it mean climbing a hill, jumping a gully, whatever. There was no 'pilot.' (I happened to take his class last year, so I learned a bit about it.) Also, look at satellites. There is no room of 100 people controlling all of the satellite trajectories and altitudes and such. This is all computer controlled. There may be a few computer operators in a control room that takes care of 100 or so satellites, but I would expect that in any autonomous system, including a possible autonomous airplane, there would still be human overseers in some sort of control room for perhaps a whole fleet of jets.
I can't talk more about this now because I have to run to class. I'll jump back in later.
Aaron atp From United States of America, joined Mar 2000, 533 posts, RR: 2 Reply 17, posted (13 years 3 weeks 3 days 13 hours ago) and read 1402 times:
you're right DLX, the media blows things like the mars mission way out of proportion, and they give very little coverage when other milestones occur.
In addition to the things you mentioned the heating and cooling cycles and radiation spacecraft must endure is another phenominal feat that aren't dealt with in aircraft. Aircraft have less than a 100 C° temp range (-60° to +40°). While spacecraft go from nearly absolute zero to several hundred degrees. Getting anything to survive just solar radiation is a feat in itself. Sure, aircraft do have vibration that loosens things up, but spacecraft have tiny grains of "dust" hitting them at several thousand miles per hour, which would be likened to flying through volcanic ash.
I've always given the space industry a lot of slack when they make big mistakes. They play a different game than the aviation industry
FDXmech From United States of America, joined Mar 2000, 3251 posts, RR: 38 Reply 18, posted (13 years 3 weeks 3 days 13 hours ago) and read 1397 times:
First, I am an absolute fan of modern airliner technology. I work with it everyday and when I'm not working I'm constantly in the books. From previous threads I'm always the first to come to the defense of modern cockpit technology when it is being maligned as a source of reduced situational awareness. So I resent your lumping me with the anti-technology radicals.
My mistake in this thread was trying to come to defense of pilots (rare thing for a mechanic) who I felt were being marginalized in their profession (eg: the a/p
does all the thinking, the pilot & dog analogy, etc).
I'm not disputing the fact that technology is progressing at an exponential rate and nor am I against it. Your predictions of an autonomous cockpit are probably right on the money. My argument (not with you) is the generalization that a pilotless cockpit is no big deal and that navigating through the air is simpler then navigating the roads.
747-600X From United States of America, joined Jan 2000, 2742 posts, RR: 17 Reply 20, posted (13 years 3 weeks 3 days 11 hours ago) and read 1385 times:
There are a LOT of remote controlled aircraft, none used for passengers obviously, but on advanced jets (A320s, 777s) the pilot does almost nothing unless something goes awry. He just has to sit there and let the airplane fly and land itself. With a few more computer programs, these aircraft could very easily handle an entire flight entirely by themselves, though I don't think there are any, as yet, capable of this. I wouldn't mind in the slightest. Automatic trains don't have people running them half the time, though of course someone is always watching over, and they are now experimenting in cars that drive themselves. I think automation is part of the future whether we like it or not, and its safer whether we like it or not. A computer program was tested on military jets to handle a tragic situation: the loss of a wing. A pilot in a simulator was completely incapable of recovering the aircraft while the computer was able to calculate enough variables at once to bring the aircraft safely in. I'm not fond of loosing the personal touches of flight, but it is inevitable, I think, that the crew is a non-essential backup.
"Mental health is reality at all cost." -- M. Scott Peck, 'The Road Less Traveled'