jimbobjoe
Posts: 450
Joined: Mon Oct 08, 2001 2:04 pm

Economist Article: Pilotless Aircraft

Sun Dec 22, 2002 8:05 pm

http://www.economist.com/science/displayStory.cfm?story_id=1487553

Haven't seen it yet discussed on the boards, but it goes through the issues involved in having pilotless aircraft, and predictions that claim that we shall be flying on them as early as 2030.

Discuss.  Smile/happy/getting dizzy
 
backfire
Posts: 3467
Joined: Fri Oct 06, 2006 8:01 am

RE: Economist Article: Pilotless Aircraft

Sun Dec 22, 2002 8:08 pm

The technology for pilotless commercial aircraft is already available today. It's the psychological aspects you have to overcome. People are nervous enough about flying as it is - it's this fear, not the technology, which will prove the greatest problem to overcome.
 
donder10
Posts: 6944
Joined: Sun Oct 21, 2001 5:29 am

RE: Economist Article: Pilotless Aircraft

Sun Dec 22, 2002 9:36 pm

Today's planes are rather similar!
 
MD88Captain
Posts: 1224
Joined: Wed Nov 14, 2001 9:50 am

RE: Economist Article: Pilotless Aircraft

Sun Dec 22, 2002 10:43 pm

Does anyone know how many pilotless Predators have been lost? 1/3 of all manufactured. 33%. Some to enemy fire, but a huge percentage just crashed due to technical difficulty. I think I will make it to retirement before getting replaced by Mr. Roboto.
 
Klaus
Posts: 20578
Joined: Wed Jul 11, 2001 7:41 am

RE: Economist Article: Pilotless Aircraft

Sun Dec 22, 2002 10:53 pm

The pilots are still needed for any kind of emergency and other unexpected situations where common sense and complex background knowledge are required.

Today´s software isn´t up to that kind of challenge, yet. Nor will it be within the foreseeable future.
 
jwenting
Posts: 9973
Joined: Mon Apr 23, 2001 10:12 pm

RE: Economist Article: Pilotless Aircraft

Sun Dec 22, 2002 11:29 pm

The RAF used pilotless aircraft in WW2 as targetdrones.

Most predator drones that were lost were lost when the radiolink failed with the base station I think. That can happen, and the drone may well crash when that happens (or go into predictable behaveour making it easy to shoot down) or may even have a self-destruct function.
I wish I were flying
 
User avatar
apuneger
Posts: 2964
Joined: Fri Sep 08, 2000 2:43 pm

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 12:28 am

Interesting topic. Now although it could save alot to the airlines and such, I don't thing we'll ever see a pax aircraft being flown without at least one pilot, preferably three...At least not for many many decades to come.

Why? Well, the reason is really simple: cost would go down (no more pilots to pay, but revenues would aswell, since there would be not one passenger left to fly to his/her destination. Let's face it, who would want to fly on an aircraft that is being flown by a computer? I know I wouldn't. The risks are just too great nowadays. Suppose something goes wrong, I'd feel more safe when there's still at least one pilot to save the whole plane and pax. As said by someone over here already, it's a psychologic thing...

Ivan
Ivan Coninx - Brussels Aviation Photography
 
backfire
Posts: 3467
Joined: Fri Oct 06, 2006 8:01 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 12:58 am

Let's face it, who would want to fly on an aircraft that is being flown by a computer? I know I wouldn't.

I'm tempted to ask what you think a flight management system, autopilot and autoland are, if they're not computers that spend their time flying the aircraft for you?

But I'm sure you didn't quite mean it that way.  Big grin
 
palebird
Posts: 75
Joined: Sun Jan 28, 2001 1:41 pm

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 2:08 am

You actually believe that pilots are flying your aircraft today? Have I got news for you. Pilotless travel is a lot closer than you think.99% of the airmiles you put on flying in the newer birds are computer flown miles.
 
User avatar
yyz717
Posts: 15689
Joined: Mon Sep 10, 2001 12:26 pm

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 3:00 am

The cost of air travel would plummet if the overpaid pilot work force could be eliminated.
I dumped at the gybe mark in strong winds when I looked up at a Porter Q400 on finals. Can't stop spotting.
 
User avatar
apuneger
Posts: 2964
Joined: Fri Sep 08, 2000 2:43 pm

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 3:24 am

OK guys, I meant ONLY by computers, as in 'no pilot on board whatsoever'... Of course I know the importance of computer systems in todays aircraft (autopilot, FMS,...).

Ivan
Ivan Coninx - Brussels Aviation Photography
 
FDXmech
Posts: 3219
Joined: Sun Mar 19, 2000 9:48 pm

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 3:37 am

Why not start smaller, let's say driverless bus's or even better, unmanned 18 wheel gasoline tractor trailers on the highways.
You're only as good as your last departure.
 
captaingomes
Posts: 6251
Joined: Tue Feb 13, 2001 1:33 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 4:59 am

Yyz717, I agree that pilots at some airlines are overpaid (i.e. United), but don't believe the costs are that significant to the airlines. With increased automation, I feel that pilots will become less significant to airlines in the future, and might even require less training in the future. That means airlines might push to decrease pilot salaries into the future.

Think of the terrorist aspect of having pilot-less airplanes? All they need to do is hack into the computers controlling the aircraft in the air, and fly them anywhere they want!
"it's kind of like an Airbus, it's an engineering marvel, but there's no sense of passion" -- J. Clarkson re: Coxster
 
User avatar
clickhappy
Posts: 9042
Joined: Thu Sep 13, 2001 12:10 pm

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 5:06 am

Pilots are overpaid (not my opinion) until they save your ass from dying. Wanna trust a computer with your life? I don't.

And, if the planes were riding some sort of radio beam, could it be hacked? Perhaps the next generation of Al-Quesadia will be computer nerds trying to crash airliners.
 
avroarrow
Posts: 804
Joined: Wed Sep 19, 2001 10:40 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 5:29 am

Well time for my .02 I guess. I would feel nervous about flying in a pilotless aircraft, just for that one in a million chance that the system might screw up. And if I read the article properly they mention having ground monitors for the aircraft to deal with ATC anyway, so why not have a guy in the cockpit instead if you are going to pay someone to do this. Many other risks include unforeseen emergencies, hijackings (which would still be possible, but in an entirely different way than we know now, likely by a computer virus). You could automate ATC as well so that all movement could be controlled by the tower directly, but that would necessitate the elimination of all piloted flights to ensure compliance (avoidance of human error), and speaking as a Cessna pilot I think that would be hugely unfair. The benefits of having a trained crew go on for much longer than I have the ambition to type. Of course history could prove me wrong....  Smile/happy/getting dizzy
Give me a mile of road and I can take you a mile. Give me a mile of runway and I can show you the world.
 
MD88Captain
Posts: 1224
Joined: Wed Nov 14, 2001 9:50 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 6:23 am

This is an absolutely stupid thread. It just shows me how little knowledge is out there concerning aviation. Very little knowledge. Almost nothing. You have no idea how incredibly busy a flight deck becomes in the departure and approach phases of flight. Listen to an approach frequency in New York center for 5 minutes. Listen to a controller giving 17 vectors in 60 seconds without even getting a readback. 17 pilots have to correctly and precisely respond to their command or aircraft start running into each other. How can aircraft be pilotless if ATC isn't controllerless? It is a dynamic and sometimes an almost frantic environment. Ever heard a controller screaming for an immediate turn only to see an airplane go zinging past? I have.

Ever been flying a Cat 2 or 3 and broken out 70' right of centerline? Trust the computers you say. If I had trusted the computer that day I'd have been a statistic.

Ever had the autopilot just disengage for no apparent reason? It happens maybe a dozen times a year to me. What happens when there's no pilot there to take over? Smoking hole is what happens.

Every had all the displays go blank? That is a complete and total nav instrument failure? Yes, it happened to me in a 757 going into JFK. We were VFR with the field in sight 30 miles out. I flew it and landed it like a 172 and no one in the back knew a thing. If there had been no pilot...

I was the FO on an 88 taking off out of ATL one day when at about 400' when the cabin started filling with smoke. We declared the emergency with an immediate turn to downwind for a very tight pattern. We completed all checklists, coordinated with tower and company and were back on the ground in 4 minutes. It was from a pilot's standpoint a beautiful thing. It turned out that the Air Cycle Machine in the right pack had exploded.

You guys can be jealous of what we make (YYZ717) but the truth is you have no idea what we do. Or like the guy who thinks I fly the aircraft only 1% of the time (palebird). You guys are clueless.

 
captaingomes
Posts: 6251
Joined: Tue Feb 13, 2001 1:33 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 7:20 am

MD88Captain, you make good points, and I for one do not want to see pilot-less aircraft. I don't think the confidence will be there, and there will be a lot of accidents that could have been preventable had pilots been in the flight deck. Are pilots overpaid? Not for the most part, but the gap between the bottom and top is too large, and I don't think pilots deserve what they get in the upper end. That's the topic of another discussion. Keep in mind that I am training to be a pilot, and it's my dream to become an airline pilot. I would also never turn down the upper end of the pay scale.

This is not a stupid thread however, I don't see how you can say that. Flight is becoming more and more automated, and as new technologies come into play, it will become even more automated. This is a reality. Modern airplanes already know when a situation arises, i.e. engine failure, and will take many of the steps by itself without pilot input. Another example is when there is a rapid decompression, the aircraft will initiate the dive by itself and level off at the required altitude. This is what I heard after a SkyService A320 had a rapid decompression earlier this year. To call others clueless is nonesense just because you don't agree with their viewpoints.

"it's kind of like an Airbus, it's an engineering marvel, but there's no sense of passion" -- J. Clarkson re: Coxster
 
MD88Captain
Posts: 1224
Joined: Wed Nov 14, 2001 9:50 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 7:43 am

I am thoroughly fustrated with the attitude that pilots are just a redundant system. And those opinions are clueless. Their opinions are worthless because they have no facts. No basis. No understanding. They've never walked the walk either.

As to automation making the job easier. That is another foolish opinion held by those who are not using the technology. The technology has made the job harder and busier. I The easiest, best flying I've ever enjoyed was the stone simple 727. The most miserable aircraft and flying I've done is the thoroughly modern and incredibly automated MD11.
 
FDXmech
Posts: 3219
Joined: Sun Mar 19, 2000 9:48 pm

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 8:02 am

Another example is when there is a rapid decompression, the aircraft will initiate the dive by itself and level off at the required altitude. This is what I heard after a SkyService A320 had a rapid decompression earlier this year. To call others clueless is nonesense just because you don't agree with their viewpoints.

As I've never worked the A320, I can only speculate this is hogwash.
You're only as good as your last departure.
 
EGGD
Posts: 11880
Joined: Sat Feb 24, 2001 12:01 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 8:07 am

How is an automated aircraft supposed to land with a variable 30 knot crosswind?

Sorry, I don't believe there will ever be a pilotless aircraft, not only because technology will not be able to cope with the ever changing conditions of the planets weather, but just like driverless cars (that we have the technology for too), nobody would want to fly them because when you remove the human element, you lose that reassuring voice over the PR system of the captain telling you what route you will be taking and what the weather is like.

Don't think its gonna happen.... If it would happen, most of us would be driving cars that drive themselves, and busses and all sorts of things...
 
ovelix
Posts: 606
Joined: Mon Aug 30, 1999 12:50 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 8:34 am

Magic of today is the reality of tomorrow.

EGDD put it right in the end. By the time pilotless airliners are doing passenger flights we will be probably ride fully automated cars, trains or other vehicles. It won't be a shock.

As for the tech difficulties (30kts crosswind, ATC etc) let's see how technology will further enhance the controlling of aircraft and how the whole enviroment of air controlling will evolve.

Many times in the past people have said "Impossible. That thing will NEVER happen" and the history proved wrong the deniers.

Anyway we 're talking for 30-50 years from now. Anything could happen.

Kostas
 
backfire
Posts: 3467
Joined: Fri Oct 06, 2006 8:01 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 8:47 am

How is an automated aircraft supposed to land with a variable 30 knot crosswind?

If you think this a challenge to modern flight-control software, then I wouldn't visit an air traffic control centre if I were you.
 
backfire
Posts: 3467
Joined: Fri Oct 06, 2006 8:01 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 8:52 am

Could someone provide me with a list of all the accidents in which a computer's actions, independent of those of the flight crew, were a primary contributory element?
 
gigneil
Posts: 14133
Joined: Fri Nov 08, 2002 10:25 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 9:19 am

... and then could someone list all the accidents in which the flight crew were the primary contributory element?

Those who say it could never be done need to look outside. Saying it shouldn't be done is another matter of more merit. But don't say it could never be done.

N
 
MD88Captain
Posts: 1224
Joined: Wed Nov 14, 2001 9:50 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 9:28 am

Pilots deal with the technology everyday and see the limitations. The many limitations. When I first saw this forum I thought it was pretty neat. I did wonder why there were so few commercial pilots posting here. Threads like this make it clear. Threads that say replace the pilots they cause too accidents. I think I'll take a rest.
 
captaingomes
Posts: 6251
Joined: Tue Feb 13, 2001 1:33 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 10:37 am

MD88Captain, it's unfortunate you are taking the views that you are taking. The majority here on this thread feel that they would not want to fly on a pilotless aircraft. I am of the same opinion. You cannot eliminate the human element in the flightdeck. However, the greater debate might be, does that human necessarily have to be in the flightdeck, or could they have one human controlling a number of flights at the same time from the ground? It's all possible, although not a comforting thought in my mind.

However, if you look at statistics, they claim that most air accidents are caused by human error in one form or another. You can't hide from the facts, and you should honestly be able to discuss in this thread the pros and cons of what the author has provided. It's all part of discussion and debate, and we all learn from different viewpoints. Right?
"it's kind of like an Airbus, it's an engineering marvel, but there's no sense of passion" -- J. Clarkson re: Coxster
 
ryu2
Posts: 1545
Joined: Tue Aug 06, 2002 8:18 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 10:51 am

Some questions to think about: what if the radio link is lost, as MD88Captain and Jwenting point out? What would happen then? How would the plane become controllable? Or fuel starvation or volcanic ash, like the Air Canada or the BA plane in Indonesia, where the power is totally lost and the computer stops working? The computer can't be programmed for everything possible?

BTW MD88captain: I find your posts informative, and hope you stay. To answer your question about why there are so few commercial pilots, I think most pilots on here participate on the "tech/ops" forum on airliners.net, not this forum, because there is almost no pilot bashing on that forum.
 
BWIA 772
Posts: 1613
Joined: Sun May 12, 2002 2:33 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 11:05 am



The article rasies some interesting points. However I could only see this happening if you had robots (computer programmes) who were like "Data" on Star Trek were being made and even then you need Humans.

For those who agree or embrace this idea of pilotless aircraftsm would they like to have computers as doctors, to go into a hospital where everything is automated. Computer technology should be used to assists us not replace us.

MD88Captain made valid points that shows us why pilots should not be elminated.

Long live the computer and the pilot working hand in hand complementing each other
Eagles Soar!
 
ryu2
Posts: 1545
Joined: Tue Aug 06, 2002 8:18 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 11:22 am

There is a big difference between planes and trains/bus.

Trains and bus are different, because if it breaks down, you can send someone to assist or fix the problem on the ground. But in midair, you can't, and the plane will probably crash, since it can't stop in midair for assistance.
 
trnswrld
Posts: 1064
Joined: Sat May 22, 1999 2:19 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 12:04 pm

YyZ717:

If you think pilots are overpaid you really are clueless. I have not the slightest clue who you are or what you do, but do you realize how much time, work, and effort an airline pilot has gone through to become and maintain what they are???

And as MD88captain said, I get SOOO aggrivated when you hear people that think they know everything say that pilots dont do anything "the plane flies itself". Last summer I was in Maui with a buddy of mine who used to fly 767/757 for TWA (now MD-80 for AA). We were sitting in a van with a group of people driving to some event we had going on that day. Everyone was friendly asking him several questions about flying, and some specifics about what he does. Well the gentlemen sitting in front of us who is a hot shot cement truck driver turns around and says and quote "you guys dont do anything up there, the plane flies itself" After hearing that I started turning red in the face. Humm, lets compare; Cement truck driver vs. 767 captain. I would love to sit him in the left seat of a 767 talking to approach control coming into Chicago O'Hare  Smile/happy/getting dizzy
 
captaingomes
Posts: 6251
Joined: Tue Feb 13, 2001 1:33 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 2:04 pm

I hope none of you pilots, such as MD88Captain are thinking I'm for pilotless flightdecks. But you have to see it from the point of view of the airlines. Some might be happy to get rid of the pilots, or reducing it to a one man flight deck. This same discussion was prevalent in the days when they reduced flight deck crews from 3 to 2. Pilots screamed bloddy murder, but it was done. It seems to work well too. Airlines are always looking for ways to cut costs, and might ignore the relevant issues outlined above. The big hope is that the unions and the regulatory bodies will have enough clout and brains to say this isn't the way to go.
"it's kind of like an Airbus, it's an engineering marvel, but there's no sense of passion" -- J. Clarkson re: Coxster
 
covert
Posts: 1498
Joined: Sun Oct 21, 2001 1:02 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 3:41 pm

I'll give a million dollars to that engineer that can program a computer to taxi a 747 around Heathrow....
none
 
Klaus
Posts: 20578
Joined: Wed Jul 11, 2001 7:41 am

Covert

Mon Dec 23, 2002 3:47 pm

Covert: I'll give a million dollars to that engineer that can program a computer to taxi a 747 around Heathrow....

Done deal... unless you added the requirement of taxiing safely around the airport...  Wink/being sarcastic

That small distinction will probably buy you another few decades of waiting...  Big thumbs up
 
covert
Posts: 1498
Joined: Sun Oct 21, 2001 1:02 am

RE: Economist Article: Pilotless Aircraft

Mon Dec 23, 2002 4:07 pm

Yeah thanks Klaus, saved me a million bucks buddy....
none
 
VirginFlyer
Posts: 3880
Joined: Sun Sep 10, 2000 12:27 pm

RE: Economist Article: Pilotless Aircraft

Tue Dec 24, 2002 1:31 am

As a friend of mine said when we were talking about this earlier this year (he is a flight instructor), 'why on earth do they want to take the best part out of an aeroplane!'

Really, I think the best option, as with anything really, is not to have an extreme of something, but to find a healthy medium. I would advocate a fully automated aircraft as much as I would advocate a fully non-automated one. Sure, both would be interesting novelties, but that is about it. The best system is one that uses both humans and computers, building on the different strengths of the two, while going as far as possible to minimise or negate the individual weaknesses. Hopefully as computer programmes get better, and pilots become more comfortable, the man-machine interaction will become even more dependable.

Ryu2 raises a very salient point about the difficulties faced in aviation as opposed to ground based transport. I think this point alone makes pilots worth every cent they are paid.

V/F
"So powerful is the light of unity that it can illuminate the whole earth." - Bahá'u'lláh
 
Ikarus
Posts: 3391
Joined: Mon Jan 08, 2001 10:18 pm

RE: Economist Article: Pilotless Aircraft

Tue Dec 24, 2002 3:25 am

Actually, my opinion on the matter conflicts with most people here. I would say, either give the pilot 100% of decision power, or 0%, and let the computer be solely responsible. I keep thinking of those (few) unfortunate accidents that were caused by clashes between pilots and computers.

Pilotless travel may happen, it may not. I'd like to see it. The Economist happily points out that labour costs are the biggest costs for any airline, that the pilot work rota for long-haulers is very restricted by the need for redundancy (i.e. if a plane is delayed by XX hours, the pilot has to be exchanged - or the flight cancelled, because he would violate working time regulations otherwise) and other costs. Well, that's a nice argument, but not convincing on its own.

Then there's the pilot errors. They do happen. (The Economist gleefully points out that the majority of all accidents are CFIT - controlled flight into terrain) But does that mean a computer would do a better job? Or even cope with highly dynamic situations like emergencies?

It's difficult to assess these things. Advances in artificial intelligence (neural networks etc.) may produce more flexible, clever computer, capable of flying planes even in emergency situations. But there are some things where computers are unlikely to be much help. For example, processing of visual information. When something goes wrong at an engine, for example, the pilot can look outside and at least get a general idea of the situation (for example whether it's still attached, or whether it's burning etc.). A computer would have to rely completely on measurements and instruments and numerical data to make its decision. Which makes it more sensitive to instrument errors. A pilot can choose which instrument to trust, but how does a computer choose? All these things need to be adressed and resolved, and many more. But eventually, a computer with flying skills superior to those of any human pilot is quite feasible, in the long run.

I don't think we're 30 years away from pilotless flight, it's at least 50 years. But eventually, it may happen. And I wouldn't be too opposed, either. After all, we trust computers with driving trains, cruiseships (where the captains are there for the tourists and not for much of the actual driving these days), elevators, spaceships, and UAVs. (Just watched a documentary about the Deep Space ion-driven mission, which was an incredible achievement, IMO)

How would it work? The Economist believes a ground controller would supervise 4-5 aircraft simultaneously (i.e. 1 human pilot for 5 planes instead of 2 for 1 plane). I think air traffic control would slowly be changed from a voice-transmitting system to data, air traffic control would be largely taken over by computers, flying the planes would be done by computers communicating with ATC computers, and humans would be supervisors, both on the ground and in the air. Initially, pilots would still fly along, but without anything to do. Eventually, the lead cabin crew would be transformed to a more empowered role - i.e. able to command the computer to land (in case of air rage, for example) or declare emergency. Perhaps he/she'd even get some rudimentary flight training for taking over in case of total power loss for the computers. (But with a provision that this may only happen if the ground controller sets the system free, as provision against hijackings)

The real challenge is to prove that computers are just as safe, or safer, than human pilots. Once that is a) the case and b) proven, their application to aircraft is only a matter of time. UAVs and spaceships will be valuable tools in improving and proving the technology. Once they log enough missions without errors and prove the idea safe and feasible, the next step will be logical...

The other factor, unmentioned by the Economist, is aircraft shape. Today, planes are very traditional in structure and shape. But look at what aircraft designers are dreaming of: Sonic Cruisers and Flying Wings and other beasts. Neither is a shape that's easily flyable - they'd both have to rely on computers to make flying manageable (much like current combat aircraft are totally unflyable without computers). Now once you have to rely on computers anyway to fly the things, the step to give them total control is only logical. So, my prediction: Pilotless planes will not happen in the near future. But when plane configurations become unconventional (which would require quite some advances in materials before that's likely to happen), their time is near.

Regards

Ikarus
 
Ikarus
Posts: 3391
Joined: Mon Jan 08, 2001 10:18 pm

RE: Economist Article: Pilotless Aircraft

Tue Dec 24, 2002 9:01 am

This is an absolutely stupid thread. It just shows me how little knowledge is out there concerning aviation.

Well, someone is clearly prejudiced. Md88Captain, I strongly suspect that just as some people know little about planes and piloting, you probably know little about the true potential of computers and artificial intelligence... I'm no expert, and indeed I have far too little knowledge about the matter to be 100% certain of my arguments, but I think keeping an open mind is important. Also, while the press is often quite ignorant, I have seldomly encountered an "Economist" article that was not well-argued and based on facts and well-thought out opinions. I disagree with most of their articles these days (mainly because they're way too pro-American and pro-Bush) but I still respect the arguments they make.

I recall reading an article 4-5 years ago about an artificial intelligence, neural-network and digital-DNA based, that was used to fly some flight simulators. The idea was that hundreds of different neural networks would be tested, those that crashed the simulated plane immediately discarded, those that managed to control it for longer cross-bred, until eventually, a program capable of flying the plane would evolve. (All that based on some of the algorithms of the PC game "Creatures") After a few generations, they got programs that flew stably. A few generations later, they got programs flying the simulated fighter in a state of continued rolling motion, which baffled the programmers and designers, until they analysed the situation and realized that this was actually the most stable state of the aircraft to be in. Human pilots would not be able (or willing) to sustain this kind of flight for a given period of time. Computers have no problems. Similarly, a computer controlled fighter is limited in its G-loads only by the structure, no longer by the human inside the plane, which is why computer controlled fighter aircraft are going to happen quite soon. In 30 years' time, they'll be ruling the (combat) skies - while civil planes will take another 20 years or so to accept / adopt the technology. As for emergencies, neural networks should be able to react to them, as well as a human pilot or better (as they'd be faster). The key is not to program the computer to blindly follow some restrictive procedures, but to actually use an artificial intelligence.

Yes, I do believe commercial pilots could be gradually replaced 50 years from now. And 70 years from now, they might be a mostly extinct job. That is, if airlines and aircraft manufacturers choose the safer, logical and cost-efficient route. If they stick to sentimentalities then perhaps not.

Regards

Ikarus
 
masseybrown
Posts: 4407
Joined: Wed Dec 11, 2002 2:40 pm

RE: Economist Article: Pilotless Aircraft

Tue Dec 24, 2002 1:29 pm

I just watched an MSNBC Special on UAL flight 232, which crash-landed at Sioux City. I'm not sure a computer could do what those pilots did.

It's coming, I suppose; I just hope they give the whatever the computer system is a REAL good test on cargo first.
 
AAR90
Posts: 3140
Joined: Fri Jan 21, 2000 11:51 am

RE: Economist Article: Pilotless Aircraft

Tue Dec 24, 2002 2:02 pm

>I did wonder why there were so few commercial pilots posting here.

You'll find us in Tech/Ops, where we don't deal in the ridiculous.
*NO CARRIER* -- A Naval Aviator's worst nightmare!
 
cloudy
Posts: 1613
Joined: Sat Apr 06, 2002 3:23 pm

RE: Economist Article: Pilotless Aircraft

Tue Dec 24, 2002 2:22 pm

I am not philosophically opposed to a pilotless airplane, but those of you who say it can be done today are wrong if you mean it can be done with the safety levels people expect out of commercial aviation. Especially if you think it can be done through AI. Software can be made reliable, but it is very expensive to make software THAT reliable. And it takes a LONG time. There is a reason information technology that protects human life is allot older and more primitive than that on your home PC. I should know. I am a programmer, albeit an unnemployed programmer  Sad.

I suspect that life critical artificial intelligence is doubly susceptible to this kind of cost overrun. Even if it wasn't, it just isnt good enough yet. AI is not nearly as smart as some of its proponents would have you believe. Consider that chess is a fairly simple and well contained game - there are no unpredictable outside varriables to take into consideration. Yet no one can yet build a computer that can consistantly beet a human grand master*. Flying an aircraft in a difficult situation has allot more varriables and they are not as predicatable. Humans will be better at this for the forseable future. AI won't be able to handle much more than it does now....for a long time to come.

The advance that would make pilotless commercial flight possible would not be a super-AI, I think. It would be a very broad datalink so reliable that the possibility of its failure could be saftly discounted. And you would need some way to give the pilot on the ground full situational awareness...even the best flight simulators have shortcommings in this area. And those beasts can cost a million or two by themselves even without such a datalink. You would need....

1. Two or three of these super datalinks per aircraft, for redundancy.

2. One enhanced full-motion simulator per 3-4 aircraft. to act as a ground cockpit.

3. A super-security system to minimize the possibility of hackers getting in, protection against a physical attack on the ground stations, etc.

To do all of this might be considerably more expensive than simply having 2 pilots per aircraft. IT people have to say thing like this allot. They should be saying it even more. Sometimes you just have to tell the customer "We can do this for you eventually. But it will take a long time. It will be very expensive. Probably more expensive than doing things the old way."

Than he will believe you when you do have something to sell him that will do him some good.

A "half-way" system that would need only a single onboard pilot might come about within our lifetimes, however. The pilot would be there in case of datalink loss or to handle really unusual emergencies - the aircraft would (like some modern airplanes) be able to do fine without him. If he is incapacitated, the datalink to the ground is lost, or the AI is lost... any one of them would be able to land the plane on their own. That would be considerably simpler and cheaper, but it still would be a huge investment. I believe something similar was proposed to fight 9-11 type attacks. That proposal involved retaining two pilots and existing avionics, but adding a reliable capability for ground control to take control of a plane and land it in the event the pilot is incapacitated or terrorists take control of the cockpit.
 
FDXmech
Posts: 3219
Joined: Sun Mar 19, 2000 9:48 pm

RE: Economist Article: Pilotless Aircraft

Wed Dec 25, 2002 10:23 am

Beyond the absurd notion of a pilotless airliner, who would be accountable for its safe operation? A computer programmer, another computer etc.

Ikarus: Then there's the pilot errors. They do happen. (The Economist gleefully points out that the majority of all accidents are CFIT - controlled flight into terrain) But does that mean a computer would do a better job? Or even cope with highly dynamic situations like emergencies?

Very misleading assumption. Fatal accidents are exceedingly rare, perhaps a more telling study would determine how often routine pilot intervention & corrections occur in the other 99.99% of normal, seemingly event free flights.
You're only as good as your last departure.
 
cloudy
Posts: 1613
Joined: Sat Apr 06, 2002 3:23 pm

RE: Economist Article: Pilotless Aircraft

Thu Dec 26, 2002 12:38 pm

MasseyBrown

As to the Sioux city crash, yes, there is a computer program today that could have landed that plane saftly. Well...it would not have landed the plane itself, but it could have varried engine thrust in such a manner as to give the pilot (or autopilot) more than enough control of the airplane to make a safe landing. A NASA pilot and programmer developed it together. I remember seeing something about it in a Wings special and in AWST. It has been tested thoroughly on an F-15 and I believe it has been tested at least once on an DC-10. Better yet, no nuts and bolts need to be moved to install this capability. In computer jargon - it is a pure "software upgrade"  Smile .

But your general points, and that of FDX mech, are valid. There are allot of things a pilot does even in normal operation that are hard to program for automatically. Generally, if a system is designed to have a human in the loop, it is very hard to take him out. Since the original designers counted on having him there, you have to practically tear everything out and start all over. That is one reason most military drones (those that aren't used just for target practice like the QF-4 is) are purpose built as pilotless aircraft. You can reduce human involvement fairly easily. Yet is very hard to eliminate the human entirely without redesigning the whole control system, or even the whole aircraft, from the ground up.

As to who is responsible for the safe operation of the aircraft, that is another good point. Someone will always have to be, no doubt, but it is always more comforting to have that person in the air with you than saftly on the ground. However, this is not necessarily a show stopper. There is no reason that if the right command structure is put in place (both human and automated), all involved will have the incentives needed to insure at least the safety level we enjoy now.

However, the tech that controls military drones is still not nearly reliable enough for commercial aviation. A system that would be is a looooonnnngggg way away. It would probably cost more than it would be worth anyway, given present technology. You pays the pilots or you pays the programmers, there ain't no free lunch  Smile........

Again....

I would not be surprisedto see AI and ground based piloting allow the airlines to go down from two pilots to one on most runs. In this case, the computers act as backup for the pilot in the plane and ground based pilots can act as backups for both. This could happen sometime in the next 20 years. This system would have the added benefit of allowing ATC to seize control of an aircraft if bad guys somehow got into the cockpit, or if the pilot was drunk, etc. All this system would involve would be much more secure and redundant datalinks with the ground, as well as some improvements to existing autopilots and avionics. Going from 2 pilots to one is MUCH, MUCH easier than going from 1 to zero.


 
J32driver
Posts: 385
Joined: Fri May 05, 2000 2:55 am

RE: Economist Article: Pilotless Aircraft

Fri Dec 27, 2002 12:55 am

Please... How cheap do you think it will be to add all this technology. Then it has to be able to taxi without running into tugs, fuel trucks, busses, other airplanes. If it makes it to the runway, you have to teach it to miss the flock of geese at 200 feet that is crossing the runway centerline, and then the Cessna 172 humming around with no transponder on for your TCAS to see. Oh yeah, and don't forget about the overzealous ATC controller that tries to get you to cross an active runway when the arriving traffic is to close.

Sorry guys.... even if you can make a system that will do all of this (for cheap, because the airlines hate expensive things) then you have to make it secure.

How many computer systems out there are ABSOLUTELY 100% secure?


Thank you... I feel dumber for even taking the time to respond to this thread.
 
Ikarus
Posts: 3391
Joined: Mon Jan 08, 2001 10:18 pm

RE: Economist Article: Pilotless Aircraft

Fri Dec 27, 2002 1:09 am

How many computer systems out there are ABSOLUTELY 100% secure?


How many humans are?

Computers can replace humans when they become as safe, or safer. (Of course, that includes having 3 or 4 redundant computer systems in case one of them crashes)
 
delta-flyer
Posts: 2631
Joined: Mon Jul 30, 2001 9:47 am

RE: Economist Article: Pilotless Aircraft

Fri Dec 27, 2002 6:36 am

Never say never!

Almost everything we do today was considered "impossible" in the not-too-distant past. At one time, the "human psyche" was thought it could not tolerate speeds over 12 miles per hour.

It's only a question of time.

Pete
"In God we trust, everyone else bring data"
 
jwenting
Posts: 9973
Joined: Mon Apr 23, 2001 10:12 pm

RE: Economist Article: Pilotless Aircraft

Fri Dec 27, 2002 6:54 am

Could someone provide me with a list of all the accidents in which a computer's actions, independent of those of the flight crew, were a primary contributory element?


The first crash of the A320 demonstrator at le Bourget comes to mind. The computers were overriding the pilot command input thus causing the crash.
I do admit that the aircraft was flying a profile that is inconsistent with normal operating procedure and the computers have since been heavily upgraded (in part because of data retrieved after this accident which saw the death of (happily only) 2 people), but it can happen.

In my mind it is quite possible to have fully automated aircraft in the not-too-distant future.
From climbout to final approach computers can even now fly the aircraft unaided under most conditions (the main problem being lack of interfacing between the flight computers in the aircraft and the ATC computers on the ground, which is still handled mainly by humans).
Of course the technical hurdles are there but I am confident that those will be overcome in a decade, 20 years at most.
In 40-50 years, manned aircraft will be prohibited in controlled airspace except special corridors.
I wish I were flying
 
cloudy
Posts: 1613
Joined: Sat Apr 06, 2002 3:23 pm

RE: Economist Article: Pilotless Aircraft

Sat Jan 04, 2003 1:36 pm

"Almost everything we do today was considered "impossible" in the not-too-distant past. "

True, but some things are truly impossible. Thats what scientific laws are, they put limits on what is possible. There has been no perpetual motion machine invented yet and there never will be. This is a pretty safe bet - the one making such a bet is protected by the laws of thermodynamics.

In my own field - it thought to be impossible even in theory to devise a computer program that will reliably predict when another program will enter an infinite loop. This is somthing that humans can do relatively easily. People wasted allot of time trying to make a program that could do this before finding out it was impossible.

Its also pretty safe to say we have not discovered all universal scientific laws yet. So there are almost certainly more impossiblities that await us as technology advances. Many of these impossibilities, I suspect, lie in the realm of artificial intelligence.

Also, possible does not neccesarily mean easy. Or economically viable. Remember in the 50's and 60's we were supposed to have cities on the moon and cars that could drive themselves. The field of artificial intelligence has made such embarassing predictions many times before. Having a robot that could passibly immitate human conversation has been promised by the 70', 80's, the 90's.....Still there is no computer conversation program that can fool you into believing that you are talking to a real human. Even when you just use text.

A very good book about this is out - its called "The Emperor's New Mind" by Penrose. He says some of the things we are trying to accomplish with computers may be impossible without knowledge of physics beyond relativity and quantum mechanics. Or perhaps even impossible. Period.

Notice that I did not say that a fully automated airplane is in that category. It is not - we just need to improve its reliability enough so that it can be used in commercial aviation. It will be a long time before this can be done in an economically and socially viable way. My point is that some things ARE indeed impossible. Other things may be possible but to difficult to be worth doing in the foreseable future. There are limits to technology.
 
L-188
Posts: 29874
Joined: Wed Jul 07, 1999 11:27 am

RE: Economist Article: Pilotless Aircraft

Sat Jan 04, 2003 2:32 pm

Invariably the first generations of a new technology will be military in use and suffer a higher accident rate then later versions, and the later civilian equipment that uses the technology. You have to remember that 33% loss rate is on the Predator, which could be argued is the first fielding of a long range UAV. The kinks will be worked out by this UAV and the other members of this generation of aircraft.




OBAMA-WORST PRESIDENT EVER....Even SKOORB would be better.
 
Alessandro
Posts: 4962
Joined: Wed Sep 12, 2001 3:13 am

RE: Economist Article: Pilotless Aircraft

Sun Jan 05, 2003 1:27 am

I discussed this topic on the "other" forum, but here´s my opinions.
Sueing, imagine what kind of law suits it would be if a pilotless plane crash
due to technical failure?
So what savings would it be to have no pilots, no pilot strikes (stewardesses, ramp personell, mechanics and so on can still strike) no pilot salaries (compare that to a law suit with Boeing/Airbus and/or the affected airline responisble for several hundred or even more if it crash into a building). The bad will if a crash
occur and it can mean long time grounding of planes, it could be very costly.
So I think "Robot airlines" are long away into the future....

From New Yorqatar to Califarbia...
 
Joni
Posts: 2613
Joined: Thu Aug 10, 2000 11:05 pm

RE: Economist Article: Pilotless Aircraft

Sun Jan 05, 2003 1:47 am

I think it's common sense to have someone overseeing the various systems on an airliner. The cost of having a pilot or two in a plane with hundreds of passengers isn't that bad.

As time goes by, the systems on airliners will likely become more and more automated and the pilot's workload will likely decrease. In the end, s/he'll be comparable to an operator at a large industrial process - most of the time doing nothing, but every now and then adjusting this parameter or that.

It also adds to people's sense of security if they deem there's someone onboard who knows the plane inside out and can, if necessary, fly it "manually".