I know the glass cockpit is a remarkable achievement aeronautically, but is there such a possibility that planes could become as automatic as the subways? Because the trains know when to "depart" and stop. Is there a possibility in the next fifty years or so that planes essentially can be computerized enough to have their own flight plans (brains)?

Would you put your family aboard such an aircraft? That'd answer your question right there....

Of course, no one knows exactly.

I think there are way too many variables and eventualities in a flight for any computer to cope with. I don't think, and I hope, we humans will ever trust any computer that much.

"..I think there are way too many variables and eventualities in a flight for any computer to cope with..."

We're already doing aspects of this through FBW. What's different ? Pilots are dependent on the flight computers converting their requests into action. There's no linkage remember.

I tend to agree with SATL382G and Aloges on this one, but at the same time, imagine if you had presented the idea of modern flight to someone 100 years ago.

The idea of getting into a large metal tube with a couple hundred other people, reading a book and watching movies in total comfort at 8/10ths the speed of sound, and then landing on another continent several hours later would have seemed a LOT more ridiculous to them than any level of automation does to us.


2H4


edited for typo

[Edited 2004-12-29 02:06:26]

I would ride on a train that has no driver. Just ask all the people in New York, Chicago, Washington, D.C., London, and Tokyo. They all would do it. People 100 years ago were afraid to death of riding on trains without engineers. Eventually I believe the story of how trains developed will also be the story of how planes developed. Why not? Pilots may eventually become obsolete....however when that will happen is clearly open to debate.

Or if it will happen for that matter...but if a subway can travel all by itself, why not an airplane? They already have military drones that fly alone with a mission programmed into their systems. Drones clearly prove pilots indeed do not have to be on board an airplane in order for it to takeoff and land.

Would you put your family aboard such an aircraft? That'd answer your question right there....

Not today, but someday...


As for going beyond the glass cockpit. I think that what we are seeing in military applications will come into commercial aviation, just as before. Virtual display of the environment, with integral instrumentation, is the future. To start with, integration of outside views with instrumentation.

HUDs are the first step. Either helmet mounted (ok spectacle mounted), large display screens or retinal projection (ok that one is further our). Instead of seeing a cockpit, pilots see a virtual room where the environment is projected along with relevant data. Dangerous terrain may be highlighted in color or even flashing. Projected flight paths (own and other planes) are projected as lines. You get the idea.

For future supersonics, front windows will be a definite "maybe" given all the problems involved (droop nose) so we'll have to go to vision systems anyway.


Even further on, it's harder to speculate. Direct brain linkage. We're seeing the very beginnings of this now. Or maybe one pilot remotely supervising several autonomous aircraft, just like Global Hawk or Predator.

There is a huge difference between the automatic subways and airliners: The failure modes.

If anything unforeseen happens to the subway, then it is programmed to stop.

Do I need to say more...?

Prebennorholm, my point exactly.

Maybe computers will be able to factor in fog, thunderstorms, tail-/headwinds, diversions, engine failures, sick pax, unruly pax, turbulence and birdstrikes one day, but yours truly living in 2004 can't imagine he'd ever set foot on a non-pilot plane.

"..If anything unforeseen happens to the subway, then it is programmed to stop.

Do I need to say more...?

The computers could just pull the plane up on the side of the jetway, whilst they display a "blue screen of death". Seriously, there would need to be many layers of checking, protection, etc. We've already seen plenty of automation of functions (e.g. reduction to two man crews through replacement of the flight engineer). The pilot is in the middle, but there is a lot of automamtion/decision making being made for them. Eample how many zero-zero landings are entrusted to Auto-land/CatIIIc ? I know the crew is present, but they're monitoring. Plenty of protection (e.g. dual auto pilots, etc.) is needed before you can deploy.

Oh, and I'm in no way attepting to devalue to the skills of a pilot in this discussion. However, the whole pilot/plane relationship is changing through FBW, flight envelope protection (correct term ?), etc.

It mighten happend for a while, but we can't discount this.

Honestly.. If the programs that flew the plane were completely tested and had a lot of redundancy.. I would feel safer flying in that plane... Think it eliminates one big source of error... Humans.. If all aircraft were flown electronically by a computer system, chances are the AA crash in bell harbor wouldn't of happened, and also the crash of the jet that ran out of fuel because of a language barrier..

Would you put your family aboard such an aircraft?

I would. Absolutely, and without a second thought.

However, I can't imagine that there will ever be a single flight without at least one pilot on it to monitor the machine and make decisions in emergencies.

N

Let's see a computer that can pull off Sioux City.

"..Let's see a computer that can pull off Sioux City..."

It's a credit to the crew that managed to bring the DC-10 back and save so many lives. However, I understand the attempts to re-fly the situation in a simulator with other crew failed to produce the same result. If this is true, it shows that not all pilots can pull off a Sioux City. I think it was the combination of crew, which included the extra pilot, plus a series of discoveries (e.g. control through differential throttle application, managed through the extra pilot) that brought that plane home.

However, can we say that automation can't be "self learning/adjusting" based on situations ? This is how the DC-10 crew managed to achieve their heroic result.

The technology is already there. The RQ-4 Global Hawk can file flight plans with the FAA, depart, and land on an intercontinental basis without human intervention. It just needs to be more reliable to be rated for humans. However, I could see it happening on freighters before it happens with passenger airliners.

[Edited 2004-12-29 07:01:17]

Maybe one day perhaps.However currently it would be tough to put ones Family on a Pilotless Aircraft.
regds
MEL

Good Topic Thrust,

I definately think the pilotless cockpit is a reality some day. Prior to that, I think the first big step is going from two crew to one. I think there can be enough electronic redundancy to remove the second person. In reality, the copilots biggest asset is the human redundancy to the pilot. Not discounting any copilot or their skill, just pointing out that the true reason we haven't gone to one is redundancy.

As too beyond the glass cockpit, Starlionblue hit the nail on the head with the new generation HUDs. In conjunction with newer GPS, the systems are pretty damn cool. Boeing is testing out a bunch of neat stuff these days, a HUD that outlines the runway, an automated taxi system, new terrain alerting systems. Visual screens instead of windows could become a real possibility as well. Not just on supersonics.

You know, thinking about it, what about digitial control over all those darn circuit breakers up there. I mean the general layout of a cockpit hasn't changed much in the way of reducing all the circuit breakers up there. I am no EE, but surely there is a better way to do that these days.

Shawn

You know, thinking about it, what about digitial control over all those darn circuit breakers up there. I mean the general layout of a cockpit hasn't changed much in the way of reducing all the circuit breakers up there. I am no EE, but surely there is a better way to do that these days.

This will happen someday, but given how seldom they are touched you might as well keep them around for now.

You know, thinking about it, what about digital control over all those darn circuit breakers up there.


Great idea, but Bombardier beat you to it.  

The Global Express uses an electrical load management system that replaces the conventional circuit breaker/relay panel, and performs automatic load control.

The electrical load management system distributes, controls, and monitors power to all aircraft loads and power busses, and automatically performs all load management and electric bus switching control functions. In case of an emergency (or electrical failure), the system automatically reconfigures the electrical busses, sheds loads if necessary, and simultaneously informs the crew, keeping them in the loop.

Take a look at the large version of this photo. The "circuit breaker" control screen is visible just to the left of the yoke.


View Large View Medium

Photo © Angus Wighton




Pretty cool, huh?


2H4

2H4

Nice! Didn't know anyone had done that.

Shawn

For every 1 crash you ever hear about their were probably 1000 times that a pilot returned a airplane to The ground through a bad situation ( weather or mechanical). people should not look at pilots as the weakest link in the cockpit because the pilot is of the up most flexibility in any cockpit operation. I realize that I'm lucky to fly a airplane with DME and I have flown with tons of bone head pilots but in a world with 3 dimensions not running on rails things become infinitely more complicated.

I for one am highly skeptical that we'll ever seem humans replaced by computers on-board airliners. By the time they figure out how to do this, we'll be beaming back and forth a-la Star Trek.  

cheers,
Phil

Please do not make any comparison to military UAVs. Like the subway trains they have a much simpler failure mode.

Failure mode of the UAV is simply "crash". And replacement with a new bird.

But thinking about the psychological issue, the CAAs, pax, airlines may be the easier ones to convince that pilotless airliners are a good idea.

But the people on the ground? How could they be convinced? And they will always be a majority.

With pilotless airliners we can just forget about having airports within 500 miles distance from populated areas. People won't accept it. Which means that maybe we will have pilotless airports on the Moon long time before we get them on planet Earth.

This no-pilot stuff could prove disastrous for the aviation windshield wiper industry...

I for one am highly skeptical that we'll ever seem humans replaced by computers on-board airliners

Never say never. When the London Underground was constructed, many critics said that humans would go insane from the enclosure, the noise of passing trains, the air, etc... This just goes to prove how much we take for granted can change with time.


Please do not make any comparison to military UAVs. Like the subway trains they have a much simpler failure mode.

Failure mode of the UAV is simply "crash". And replacement with a new bird.

Granted. But UAVs are still in early days in many ways.


This no-pilot stuff could prove disastrous for the aviation windshield wiper industry...

The windshield industry would have a problem too LOL.

I think people here are examining the possibility of pilotless aircraft with two very different perspectives...

The first perspective considers the probability of this happening with today's technology and the technology of the foreseeable future. In this context, safe, pilotless commercial aircraft may be possible, but are quite improbable. This is due to our relative lack of experience with pilotless aircraft and the consumer's unfamiliarity with them.

The second perspective considers the probability of pilotless commercial aircraft that utilize technologies that have not yet been developed. Aviation is only ~100 years old. Who's to say what's possible or impossible in the next 50, 100, or 200 years?


2H4

Can an airliner fly without a pilot? The answer is yes and you don't need to wait for decades for it.The menkind has put a man on the moon and guess what?They did it all by computers which can not even open a page on the A.Net website today(this was from Discovery channel)

I can tell you about my last flight (I am not flying nowadays because of the company and acft type conversion).It was a short trip ended with a practice autoland.(Airplane did the landing itself).I only touch the yoke during the TO roll until 1000 feet and engaged the AP and the airplane did the rest itself.I am sure there is technology to make it TO itself too.

So if you think that I am not needed on that airplane only because I only manuplated the yoke for 20 seconds on the entire flight you are missing the the difference between a pilot and the "airline captain".Piloting the airplane is only a part of the job,which is constantly getting easier and safer by the advances in technology.

But there is still no computers to make decisions during positions like; passengers missing after boarding compleated and there is a slot time you have to catch,would you leave them or not?A pregnant lady claims she is fit for the flight but can not show a doctors report,would you accept her or not?Weather looks fine now but might be just in your limits upon your arrival according to TAF,would you TO or wait?What about the slot times?How about your fuel,how much extra will you take?Do you have to sacrifice any loads from the airplane to take the fuel?How will you transfer the luggage later?You diverted to alternate,will you send the pax to the hotel?How about if the weather improves?How long will it take to bring them back to the airport?Would it be better to send them to the destination with another type of transportation?Will it save the company money or cause the company to loose prestige?............This is only the start of the list and it goes on and on.

If you remove the pilot from the plane you will still need someone to preflight the airplane(you know the autopilot is so dumb that it doesnt even know which airport he is on if you don't load it to the FMC),And you need another person(s) to make the decisions I counted above.(They are everyday airline business you know)

So what is the catch of spending billions to make a pilotless plane and putting (atleast) two people instead of him to make his job?

If we ask the question from the beginning again.Can airliner fly without pilot?The real answer lies in the airline companies.They only invest money if they know that they will earn more benefits from their investment.And if removing the pilot should bring any good they should have done it long time ago.

Wing: Can an airliner fly without a pilot? The answer is yes and you don't need to wait for decades for it.The menkind has put a man on the moon and guess what?They did it all by computers which can not even open a page on the A.Net website today(this was from Discovery channel)

Neil Armstrong landed Eagle fully manually using one of the first FBW control systems in the world.

Neil had one big advantage: He was the first pilot in the world landing a vehicle with a 100% certain knowledge about weather conditions.

But back to airliners. You front seaters, you tell us that you control the first 1000 feet of the TO roll and the FMC does the rest. But that's when everything goes as planned.

You also tell us that you lose a gallon of sweat for every hour in the sim. Please tell us, what are your doing in the sim? Are you mostly flying a perfect schedule according to plans? Hardly. Or is there some nasty sim operator who constantly cripples your virtual a/c, plays havoc with the weather while filling your ears with the most stupid ATC commands you have ever heard? I guess so.

All the sweat production in the sim, is that really relevant? Couldn't you ask the computers to do that? Maybe some day. At least if you can guarantee that Bill Gates haven't even seen one single code line in the software.

Well, as I said before, as a man, living on the ground, I won't allow you to leave it all up to computers. And I decide - because we are many - about six billion.

Neil Armstrong landed Eagle fully manually using one of the first FBW control systems in the world.

Whether it was first or not is of course open to debate, although for sure it was a pioneer. The Arrow was earlier I believe.


All you luddites can resist all you want. Someday there will be pilotless passenger planes. I think that someday people will be amazed that we dared to fly without computers controlling everything. Just look at how amazed we are today that early fliers went up with only the most rudimentary of instruments, no weather forecast, nothing but basic navigational aids...

Long posts always boring to read I guess thats why I couldn't make myself clear enough.Here is the summary of what I was trying to say;

Technology for the pilotless flying airplane is available even now,but there is no logic behind it nor it is feasible.(And I am not even talking about the emergency situations)

There's also a very important concept in regard of automates and computers.

Since the beginning of the 'computer era' if I can call it that way, there's always the hype that computers could become intelligent machines that could think by themselves. While the new technology has progressed many times this last decade, and computers are more and more able to analyze a said situation, and take actions in regard of it, they cannot think by themselves. All they can do is merely choose a solution from a pool of actions that were programmed by humans. Yes, they can do if much faster, in a much reliable way than humans, but they still only do what humans programmed them to do.

For most of the tasks involved in flying an airplane in a standard situation, they can pretty much do everything, from engine management to CATIII landings. They could probably even taxi by themselves, heck, computers can drive a pool of cars on the highway without any human intervention. But that's because humans programmed them for those tasks, and against known faults, with associated actions. What computers lacks is the capacity to learn from an unknow situation, and you can't revert to a default action in this case, unlike the train (if it can't find a way to solve the problem, then stop). Subways only have 2 basics functions : forward of reverse. Planes have some mores to think of  .

I'm sure that as the technology evolves, the computers will take more and more of a place, replacing those basics and repetitive tasks. But in the end, there will always be a need for a human action in the case of an unknow situation. We, humans, can learn in a way that technology can't.

Tron will not happen  

Since the beginning of the 'computer era' if I can call it that way, there's always the hype that computers could become intelligent machines that could think by themselves. While the new technology has progressed many times this last decade, and computers are more and more able to analyze a said situation, and take actions in regard of it, they cannot think by themselves. All they can do is merely choose a solution from a pool of actions that were programmed by humans. Yes, they can do if much faster, in a much reliable way than humans, but they still only do what humans programmed them to do.

I'm sure that as the technology evolves, the computers will take more and more of a place, replacing those basics and repetitive tasks. But in the end, there will always be a need for a human action in the case of an unknow situation. We, humans, can learn in a way that technology can't.

Granted, but this doesn't mean they will never become sentient. Also, sentience is not really a requirement for controlling a vehicle. Autopilots and FMCs are not sentient, for example.

And finally there is the question of defining sentience. Even if the AI cannot pass a Turing Test it may be "smart" enough to learn and act autonomously with significant freedom.

Let's see a computer that can pull off Sioux City.

The majority of aircraft accidents are caused by so-called "pilot error". The vast majority of these "pilot error" accidents are caused by the flight crew failing to follow the standard operating procedures or operating in contradiction of the aircraft flight manual. (AFM)

For every Sioux City, there are at least 10 accidents that could have been avoided if a the human error factor were removed.

I'm not saying technology is quite ready to take over from pilots but it certainly is the in the realm of possibility for aircraft to safely fly themselves.

Pilots on the other hand continue to price themselves out of the market....

Pilots on the other hand continue to price themselves out of the market....

That'll make you popular on this board  

Pilots are actually not really paid that much compared to the huge investment they put it. Most pilots don't make a lot of money until the tail end of their careers.


Anyway I agree with you on the other point. Yes, Sioux City was a tremendous feat of airmanship, but what about all those other accidents. Who knows what the future may hold.

was reading an aircraft business magazine recently and one of the big freight operators (DHL or FedEx, I cannot remember which) is seriously working on automated freighter aircraft (they say within ten years)

FAA is modifying the regs regarding UAVs in commercial/controlled airspace and NASA is well advanced on the "Highway in the sky" concept.

Its only a matter of time unmanned (or maybe observer on board) flight will arrive. First the freights then the self loading cargo aircraft.

Its down to economics and business. Computers do not have flight and duty time issues, recurrent training, strikes, pay demands, etc ............. but at least the engineers and technicans will still have work to do to keep them flying.

Will the passengers complain? Depends how its marketed. Most people do not know how many people fly an aircraft anyway (be it a three man crew, replacement crew for long haul flights etc) and handled by the PR people correctly passengers may perfer a computer.

Where have the old jobs have flight engineer, navigator, radio operator gone....... automated thats where. Its the price of progress.

Its down to economics and business. Computers do not have flight and duty time issues, recurrent training, strikes, pay demands, etc ............. but at least the engineers and technicans will still have work to do to keep them flying.

Whiskeyflyer saves his job for the future while eliminating pain in the a** pilots out of the place for ever,that should be a dream come true for the maintenance people I guess   (Hey I am only joking please take no offence)

Its only a matter of time unmanned (or maybe observer on board) flight will arrive. First the freights then the self loading cargo aircraft.

Thats what I was referring in my previous post.You are putting someone to observe the progress and another one load data and make decisions.So you are removing 1 pilot and putting atleast 2 people instead of it.That is not something that the airlines would want.

I guess all of you are only looking from technological perspective.Boeing or Airbus can invest couple of billions and make the pilotless airplane available in no time.But to make it fly you need to install a lot of nav aids and updating equipment and airport improvement.And yet you can not make it 100% operational everyday.A simple ILS or VOR maintenance could cause delays even on a CAVOK day while a pilot could land the airplane without any additional help.

All in all I am not saying pilotless flying it is not possible,I am saying it is not logical and feasable

PS: I wish all of you a very happiness ,health and success in the new year.WING

All in all I am not saying pilotless flying it is not possible,I am saying it is not logical and feasable

Thats what many people thought when airplanes were first proposed.....

What exactly makes it illogical or not feasible?

It is currently economically illogical. It is technologically quite feasible.

I disagree. I believe it is:

1. Economically logical
2. Technologically feasible
3. Politically infeasible

Economically logical

We could argue about this one all night. Considering the investment required versus the current system of pilots, I think the returns would be deep into the future on this one.

Someday...

they cannot think by themselves

This is not correct. One of the largest areas of research currently going on in the computer science field is with computers that do just this. Search for "machine learning" or even the more traditional "artificial inteliligence" and you may be amazed at what computers can do. They no longer must choose from a set of pre-programmed options, that's for sure.

I'm avoiding the other debate because it's all conjecture at this stage.

they cannot think by themselves

This is not correct. One of the largest areas of research currently going on in the computer science field is with computers that do just this. Search for "machine learning" or even the more traditional "artificial inteliligence" and you may be amazed at what computers can do. They no longer must choose from a set of pre-programmed options, that's for sure.

I'm avoiding the other debate because it's all conjecture at this stage.


The issue soon moves into philosophy. Learning machines exist today. But they are pretty certainly non-sentient.

I would like to just add this ANYTHING IS POSSIBLE if we continue to evolve computer at the rate we are currently in !!!!!!!!!50!!!!!!!!! years anything is possible much more than just planes but everything could be computerized look at how good it is we are all sitting at one right now and tell me in the room your in now there is not at least 5 electronic items. Check mate!

I just hope we don't go to far...

Another point is how many computers to build the planes in the first place currently... compared to 20 years ago.

Aaron J Nicoli

We are working on unmanned civil transport aircraft at the moment, in the 30-40 year timeframe.

The main difficulties are not the technological aspects, but the procedural, political aspects and the attitudes of the paying customers. As an off-shoot of this work, I asked A.Net to poll users as to whether they would travel on such an aircraft. The results so far are fairly predictable, but it will be interesting to conduct the same poll say, every five years to see how attitudes change. The poll can be found at:

https://www.airliners.net/discussions/polls/index.main?id=68

One of the major point of concern is how we go through a transitional phase, where there will be both manned and unmanned aircraft in the same airspace.

Once the programme is completed, I'll try and get permission to post the key highlights here.

Eample how many zero-zero landings are entrusted to Auto-land/CatIIIc?

Probably fewer than you're thinking. Most autoland capable acft require 600 feet visibility as a minimum (limited number can go as low as 300 feet). In the near future you'll see fewer autolands as hand-flown HUD approaches have proven to be more accurate and significantly less costly to maintain. In the long term who knows, but I will not get on an airplane where the final decisions are being made somewhere else (I'm biased I know). And NO, I do not get on "no-operator-aboard trains" either. I want the guy/gal making the life or death decisions to be making that decision about his/her own life as well.

And NO, I do not get on "no-operator-aboard trains" either.

Not even the Airtrain at EWR or the trains at MCO and ATL? Is it even possible to transfer at MCO without using the train?

Oh and don't look now, but your nuclear power station is probably also being controlled by a computer. And ATC is a case of control without "being there".

ATC isn't "in control" of the aircraft. The final authority for every action on an aircraft is the PIC. There have been many times in my career where we couldn't comply with an ATC request due to safety reasons.

Technology has removed the navigator, flight engineer and radio operator seat from aircraft even though the operational capabilities have increased dramatically.

Its only a matter of time before we see single operator aircraft where a caretaker supervises the airplane's operation.

NOBODY has mentioned the possibility of having cockpits with only ONE pilot...

I'm only against this because computers might have my future job one day. 

I think I will live long enough to see pilotless 747 sized cargo aircraft though.