DAL763ER
Topic Author
Posts: 524
Joined: Thu Oct 02, 2008 7:20 pm

Planes Can Fly Themselves. BS!

Mon May 27, 2013 11:49 am

I was talking to my dad last night about the BA event at LHR this weekend. He's an ATC, I'm an aviation nut and a programmer. So what occurred to us was that while many people say "In x years, ATC and pilots won't be needed. Planes can fly and guide themselves.". While true in the ideal scenario where the plane doesn't need to do anything but fly, I would like to know what planes could do at this stage on their own in an emergency event.

Not only the plane involved, of course, but also the planes around him. Like the plane at LHR or any plane that needs to return immediately, it would need to coordinate with other planes, airports etc on its own so that it can land back safely. Then the planes that are holding because of the emergency event need to hold/divert, make sure they don't run out of fuel etc.

It all seems a very very complicated scenario, reason why I believe that the prospect of no, or just one, pilot and no ATC is very remote at this stage, despite what "experts" say.

I'm aware this has been discussed in some ways in the past, but I thought I'd give it a bit of a twist since some people that think about planes flying themselves forget that sometimes stuff hits the fan and even today a computer might not be able to handle it. Heck, even our iPhones crash, let alone a multi-million dollar plane with hundreds of people on it.
 
aloges
Posts: 14849
Joined: Tue Jan 17, 2006 3:38 am

RE: Planes Can Fly Themselves. BS!

Mon May 27, 2013 1:01 pm

Quoting DAL763ER (Thread starter):
Planes Can Fly Themselves. BS!

   They may be able to do that as long as everything is going well. But as soon as something goes wrong, which it does all the time, you need a human being that is able to think outside of the box.
Don't cry because it's over, smile because it happened.
 
Clydenairways
Posts: 1239
Joined: Thu Jan 11, 2007 8:27 am

RE: Planes Can Fly Themselves. BS!

Mon May 27, 2013 1:47 pm

Lots of threads on here about this before and most come to the same conclusion that it's not technology that will dictate this but rather public acceptance.
I think it will be at least 50+ years before we see the traveling public accept pilotless aircraft. We may see single pilot ops before this.
 
26point2
Posts: 1080
Joined: Wed Mar 03, 2010 6:01 am

RE: Planes Can Fly Themselves. BS!

Mon May 27, 2013 3:31 pm

Most would not accept flying on a pilot-less plane regardless of technology. Would you? Perhaps cargo ops...at best.
 
AWACSooner
Posts: 2486
Joined: Tue Jan 22, 2008 12:35 am

RE: Planes Can Fly Themselves. BS!

Mon May 27, 2013 3:54 pm

Quoting DAL763ER (Thread starter):
So what occurred to us was that while many people say "In x years, ATC and pilots won't be needed. Planes can fly and guide themselves."

These many people wouldn't be relatives of MoL, would they?
 
2175301
Posts: 1580
Joined: Wed May 16, 2007 11:19 am

RE: Planes Can Fly Themselves. BS!

Mon May 27, 2013 6:57 pm

Quoting DAL763ER (Thread starter):

I'm aware this has been discussed in some ways in the past, but I thought I'd give it a bit of a twist since some people that think about planes flying themselves forget that sometimes stuff hits the fan and even today a computer might not be able to handle it. Heck, even our iPhones crash, let alone a multi-million dollar plane with hundreds of people on it.

The issue is not will the automated controls fail and crash a plane with people on-board. Everyone knows that will in fact happen. The question is will it happen more or less often than pilots making mistakes and crashing the plane with people on-board. If automatic controls can be shown to have fewer crashes - then the pilots days are numbered as an active pilot; because the pilots will not be able to claim they make the planes safer. Insurance rates will also reflect a lower cost for automated flights versus piloted flights.

Now I am not saying that we are to the point where automated controls are ready to take over yet. But, I will tell you that from oil refineries to chemical process plants with dangerous/hazardous chemicals. to nuclear power plants - that the automated controls are far more reliable and safer than are manual controls with well trained operators.

I am sure that NASA and its equivalent in other countries do in fact have "experimentation" automated controls for aircraft that are in fact more reliable and safer than pilots. How soon can a real world system and planes be built to use that technology? Likely 20 years if we decide to do it.

Have a great day,
 
User avatar
lightsaber
Moderator
Posts: 18421
Joined: Wed Jan 19, 2005 10:55 pm

RE: Planes Can Fly Themselves. BS!

Mon May 27, 2013 7:03 pm

Quoting clydenairways (Reply 2):
Lots of threads on here about this before and most come to the same conclusion that it's not technology that will dictate this but rather public acceptance.
I think it will be at least 50+ years before we see the traveling public accept pilotless aircraft. We may see single pilot ops before this.

  

We've discussed this many times here on a.net and the same conclusions have been reached. Perhaps with the co-pilot in regional centers acting as needed.

Quoting 2175301 (Reply 5):
am sure that NASA and its equivalent in other countries do in fact have "experimentation" automated controls for aircraft that are in fact more reliable and safer than pilots.

NASA is no where near the leading edge. Every UAV equips a business jet with the control systems far before the first prototype of that UAV flies. The business jet still has two pilots, but they will hand over control to the computers as the pilots are there just for backup/shackdown of the pilotless system.

Those systems are becoming more and more advanced as the military wishes to transit UAVs through commercial airspace without the current expense of a chase plane. So they are working on the systems. And I've seen aircraft saved by the automatic systems that would *never* have survived if only a human was at the controls.

When will be public acceptance. First will be smaller aircraft (say 50 seaters and less). Then maybe cargo... But to act as if pilots couldn't be more efficient?   


Lightsaber
IM messages to mods on warnings and bans will be ignored and nasty ones will result in a ban.
 
2175301
Posts: 1580
Joined: Wed May 16, 2007 11:19 am

RE: Planes Can Fly Themselves. BS!

Mon May 27, 2013 7:09 pm

Further clarification on the adoption of automated technology vs lives. Antilock brakes on automobiles are a very relevant example.

Anti-lock brakes do not produce the best outcome in all situations. The highway/insurance industries identified more than 20 different road conditions where emergency breaking occurs with common loss of control of the vehicle and accidents.

My recollection is that in 17 of those cases anti-lock brakes result in better or the same level of control with fewer significant accidents. In the other cases anti-lock brakes produce worse results and more severe accidents (examples - dry pavement and straight line stopping - locking the wheels do in fact stop the car faster and loss of control is rare; another example - snow on ice: locking the wheels builds up a snow pile in front of the wheels slowing the car faster).

However, when all is said and done - anti-lock brakes on average reduce severe accidents, severe injuries, and deaths by about 2/3 over cars without anti-lock brakes (even if they make some accidents worse). Thus, anti-lock brakes are pretty much standard now (and may in fact be a required safety feature by this time).

Stability control systems have a similar reduction (but make some situations worse).

I am sure that the same logic will in fact be applied to automated flight control in the end - if nations build the infrastructure needed to implement the systems.

Have a great day,
 
Passedv1
Posts: 649
Joined: Sun Oct 07, 2012 3:40 am

RE: Planes Can Fly Themselves. BS!

Mon May 27, 2013 8:04 pm

In 50 years who knows.

Much closer than that and I think the only people having this discussion are engineers.

Before there are pilotless airplanes, engineers will invent an auto-throtle system that can keep up with moderate mountain-wave and an auto-land system that can keep up with a gust factor of more than 10 knots and there needs to be an auto land system developed that doesn't only work at only 20% of the world airports. The auto land system would have to be developed that didn't require large portions of the airport to be protected while auto-land operations were happening.

Even assuming you had a capable airplane with proven/accepted technology today, it would be a MINIMUM of 10-15 years before it was ever allowed by the FAA. Alaska Airlines started the process of getting rid of paper approach charts 4 years ago... a simple electronics representation of the exact thing we were using before...and it is STILL not approved.

As far as the pilot bashers on here saying saying the bar is lower because of the pilots..I say, the bar is A LOT higher than you think , I say the bar is A LOT higher than you think because almost all pilot "saves" are saves because nothing happened, they don't make the news , there are no statistics. These would start showing up as soon as you removed the pilots.
 
Mir
Posts: 19491
Joined: Mon Jan 05, 2004 3:55 am

RE: Planes Can Fly Themselves. BS!

Mon May 27, 2013 8:47 pm

Quoting clydenairways (Reply 2):
I think it will be at least 50+ years before we see the traveling public accept pilotless aircraft. We may see single pilot ops before this.

I don't think we'll ever have single-pilot airliners. We may have airliners with one pilot on them, but they'll be designed to be fully automated and the pilot will just be there for feel-good purposes. The infrastructure that would be needed to maintain a single-pilot operation with appropriate levels of safety and redundancy would likely cancel out the cost savings from not having the other pilot around. I'd suspect that we'll go right from two pilots to none (at least none required).

-Mir
7 billion, one nation, imagination...it's a beautiful day
 
spink
Posts: 317
Joined: Tue Aug 09, 2005 5:58 pm

RE: Planes Can Fly Themselves. BS!

Mon May 27, 2013 8:53 pm

From a technology perspective, we are actually already there. From an infrastructure perspective, we have a ways to go. The technology is certainly already there to make a pilot-less plane and has been for several year. Most of the actual technologies needed are much simpler than that required for automated cars due to both the limited amount of traffic and already existing system of over all control of flight constraints being outside of the hands of pilots (aka ATC is mostly in control of where planes are and how they go).

From an infrastructure perspective, a lot more work is required on. The ATC and aviation infrastructure for the most part is still trapped back in the 60s and 70s.
 
User avatar
lightsaber
Moderator
Posts: 18421
Joined: Wed Jan 19, 2005 10:55 pm

RE: Planes Can Fly Themselves. BS!

Mon May 27, 2013 11:49 pm

Quoting 2175301 (Reply 7):
My recollection is that in 17 of those cases anti-lock brakes result in better or the same level of control with fewer significant accidents. In the other cases anti-lock brakes produce worse results and more severe accidents

Only with a top driver at peak alertness. Porsche used to argue ABS had too many faults. Until every 944 in a 944 vs. Corvette race spun off the track in the rain. It was an endurance race. The race car driver was better than ABS until fatigue set in. Now put grandma behind the wheel and ABS is superior every time.

The same is true of aircraft. An alert pilot is superior to the automation. No doubt about that. But the automation can reduce the pilot's workload so significantly that they can then do their job.

I do not see pilot-less passenger aircraft. But after the Google cars?   
Time...

Quoting spink (Reply 10):
The ATC and aviation infrastructure for the most part is still trapped back in the 60s and 70s.

   That will have to be upgraded.


Lightsaber
IM messages to mods on warnings and bans will be ignored and nasty ones will result in a ban.
 
spink
Posts: 317
Joined: Tue Aug 09, 2005 5:58 pm

RE: Planes Can Fly Themselves. BS!

Mon May 27, 2013 11:55 pm

Quoting lightsaber (Reply 11):
The same is true of aircraft. An alert pilot is superior to the automation. No doubt about that. But the automation can reduce the pilot's workload so significantly that they can then do their job.

I do not see pilot-less passenger aircraft. But after the Google cars?
Time...

And I highly doubt that any pilot is fully alert for a full 8 hour shift. I doubt that anyone is in any industry. That is why we have automation and breaks.

Google car is nothing. The autonomous overland trucks are much more impressive! But both are more complicated than pilot-less aircraft simply because they have to deal with a chaotic dynamic environment which doesn't really exist with aircraft.
 
Mir
Posts: 19491
Joined: Mon Jan 05, 2004 3:55 am

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 1:54 am

Quoting spink (Reply 12):
Google car is nothing. The autonomous overland trucks are much more impressive! But both are more complicated than pilot-less aircraft simply because they have to deal with a chaotic dynamic environment which doesn't really exist with aircraft.

But their redundancy level doesn't need to be as high. If something goes wrong, they can just come to a stop. Not as easy to do that in an airplane.

-Mir
7 billion, one nation, imagination...it's a beautiful day
 
Alnicocunife
Posts: 197
Joined: Sat Mar 26, 2011 3:32 pm

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 1:55 am

Elevator operators They will never go away, how will people know what floor to stop on.

Why are there still train or subway drivers? Go, forward or backward and stop. Yet we still have them and there are still accidents.

It might be a while but it will come and safety will be the reason why.
 
spink
Posts: 317
Joined: Tue Aug 09, 2005 5:58 pm

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 2:02 am

Quoting Mir (Reply 13):
But their redundancy level doesn't need to be as high. If something goes wrong, they can just come to a stop. Not as easy to do that in an airplane.

3x computer redundancy is already a pretty solved problem. The redundancy requirements are not stopping pilot-less planes. It is much more of an infrastructure problem at airports and ATC. The reality is that airline pilots while highly trained are 99% glorified mechanical devices taking orders from ATC via radio and imputing that into flight computers. Once ATC and digitally send order to flight computers the pilots will be primarily reduced to looking for hazards, something that is primarily already done by computer.

So the reality is that all the technology is there for pilot-less commercial flying. Public acceptance and confidence along with infrastructure are the only things holding it back.
 
Mir
Posts: 19491
Joined: Mon Jan 05, 2004 3:55 am

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 2:15 am

Quoting spink (Reply 15):
The reality is that airline pilots while highly trained are 99% glorified mechanical devices taking orders from ATC via radio and imputing that into flight computers.

You make it seem like ATC is flying the planes, but that's not the case. ATC directs traffic, nothing more.

-Mir
7 billion, one nation, imagination...it's a beautiful day
 
Bobloblaw
Posts: 2406
Joined: Fri Jan 13, 2012 1:15 pm

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 2:18 am

Quoting clydenairways (Reply 2):
I think it will be at least 50+ years before we see the traveling public accept pilotless aircraft.

Cargo would be a good test market before passengers.
 
planemaker
Posts: 5411
Joined: Mon Aug 25, 2003 12:53 pm

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 3:02 am

As a few have said already, this topic has been discussed a couple of times but their have been many more threads on SP Ops.

The key technology piece that is missing is not infrastructure... infrastructure is on track for full deployment within 7-10 years.

The only missing "technology" is software: 4-D/Sense & Avoid. And that is expected to be deployable within 7 years.

Quoting Mir (Reply 16):
You make it seem like ATC is flying the planes, but that's not the case. ATC directs traffic, nothing more.

That is now... but pretty soon ATC will indeed "fly" the airplane if they chose/need to..
Nationalism is an infantile disease. It is the measles of mankind. - A. Einstein
 
kiwiinoz
Posts: 2000
Joined: Mon Oct 10, 2005 3:07 pm

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 3:05 am

I would accept flying on a plane that is pilot-less, providing that all elements of the aviation environment are managed through a central, integrated system, (ATC, other aircraft, emergency responses, etc)

I think the technology is pretty much there in terms of flying a plane individually. However coordinating globally a central management system is probably years away, (and may never happen)
 
ABpositive
Posts: 180
Joined: Mon Nov 07, 2005 2:36 pm

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 3:41 am

I believe we are more likely to see first a combination of single-pilot decks, greater reliance on onboard computers with groud support for emergencies (e.g. similar control centers that are used for flying drones).
I'm not sure that pilots will ever disappear. The train systems around the world have signalling and safety records superior to the aviation yet they still persist with having a train operator.
 
spink
Posts: 317
Joined: Tue Aug 09, 2005 5:58 pm

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 3:44 am

Quoting Mir (Reply 16):
You make it seem like ATC is flying the planes, but that's not the case. ATC directs traffic, nothing more.

ATC plus the planes themselves are largely doing more flying than the pilots are. The airplanes are taking off and landing on directed ATC paths, they are largely climbing, cruising, and descending on ATC control directed paths. The pilots in the cockpit are there primarily in a man-in-the-loop role. Or do you think the pilot is controlling the throttle, calculating the glide slopes, climb rates, rudders/flaps, etc during most of the flight? The only reason the pilots have to interact so much with the airplane during takeoff and landing is that the airport infrastructure largely isn't up to snuff for automated take-offs and landings, though many planes have capability in their flight control software for exactly that.

The flight crew of an airplane is doing less and less, and this isn't some new trend. In the old days, there were at least 3 people in the cockpit, one just monitoring instruments: the flight engineer. We automated that away. We will eventually automate away the co-pilot and then the pilot himself. In fact, the reality is that most pilots are not really pilots anymore but rather command and control relay officers and co-officers. About the only time they do any actual piloting that cannot simple be done by the computers already on board the aircraft is in catastrophic situations (water landing with both engines out, etc).
 
planemaker
Posts: 5411
Joined: Mon Aug 25, 2003 12:53 pm

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 3:51 am

Quoting ABpositive (Reply 20):
I believe we are more likely to see first a combination of single-pilot decks, greater reliance on onboard computers with groud support for emergencies (e.g. similar control centers that are used for flying drones).

As has been noted previously, the military could very well have the first SP Ops cargo plane... they already have tested no-pilot re-supply helicopters in Afghanistan. Furthermore, the mission profile of the Navy's X-47 will clearly demonstrate that feasibility.

Quoting ABpositive (Reply 20):
I believe we are more likely to see first a combination of single-pilot decks, greater reliance on onboard computers with groud support for emergencies (e.g. similar control centers that are used for flying drones).

FYI, there is already the "ground support" in place... at the airlines OPS' Centers and at the major OEM's that are already monitoring/interrogating and sending data to individual aircraft on a real-time basis.
Nationalism is an infantile disease. It is the measles of mankind. - A. Einstein
 
User avatar
garpd
Posts: 2507
Joined: Thu Aug 11, 2005 9:29 am

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 6:53 am

This whole Pilotless or Single Pilot idea gives me the willies.

Until a fully developed true to life Artificial Intelligence is ready to go, the whole idea is a very, very bad one.

Sure, program it right and a plane, train, car or boat can manoeuvre all on it's own. You can even program in many variables so it knows what to do in any given scenario.

Throw it outside that program and the computer becomes next to useless.

Should something not foreseen in the programming happen, the computer will either take a wobbly, crash (system crash) or decide the situation is something else and take corrective measures that could end up exacerbating the situation.

Take the BA 777 crash landing. The computer would have followed procedure to the letter and as experts have said, following established procedure would have resulted in the aircraft piling into houses on the approach path as it would have stalled out. It was the Captain's out of the box, none procedural, thinking in raising the flaps that saved everyone on that flight.

There are countless situations where out of the box thinking has helped save people from certain death, and not just in Aviation.

Computers without a true A.I. cannot think out of the box. They can only compute what they are programmed to compute.

Without a tried and tested A.I, single pilot or no pilot ops are a completely and utterly unacceptable idea. Even then, a human with experience should always be on hand as an A.I. like all computers and programs are only as good as the humans that program and code them.
arpdesign.wordpress.com
 
ZKSUJ
Posts: 6872
Joined: Tue May 18, 2004 5:15 pm

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 7:06 am

No doubt it can happen with everything working well. Once sh*t hits the fan, maybe not so good an idea.

I'd like to see the Hudson A320 land like it did with only ATC and copmputer imput. Or the Transat A330 into the Canaries, or the AC 767 at Gimli, or the BA 777 at Heathrow, or the Dash 8's that went in with a main gear up at SAS, or the Sioux City DC10 with no hydraulics, or... I think I've made my point
 
peterjohns
Posts: 176
Joined: Fri Jan 02, 2009 11:49 am

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 8:14 am

Hi guys
I´m an ATC since 25 years. Over time so much has become automated and computerised. Most of it is good and helps making things safer. Just think of TCAS - it saved the day more than once. On the other hand - it also has made some critical situations ( some might say it was even a major cause of the Überlingen accident) but overall I wouldn´t want to work without it anymore.
Datalink, Mode S, enhanced GPWS, just to mention a few more.
It is quite possible that in the next perhaps 25 years things will be operated just as the thread says. Why not? Even today
planes (Global Hawk e.g.) fly by themselves. It is possible.

Quoting garpd (Reply 23):
There are countless situations where out of the box thinking has helped save people from certain death, and not just in Aviation.

That is true. No doubt that what your saying is plausible for your examples- BUT- believe me, there are also a lot of cases where it would have been better not to have had hands on the controls...
Just think of all the CFIT´s , AF447, the Libyan A330 and countless others.

I´m sure MOL would buy a few hundred non Pilot a/c as soon as they are available!!
 
warden145
Posts: 539
Joined: Wed Aug 18, 2010 8:36 am

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 8:21 am

Quoting garpd (Reply 23):
This whole Pilotless or Single Pilot idea gives me the willies.

Until a fully developed true to life Artificial Intelligence is ready to go, the whole idea is a very, very bad one.

Sure, program it right and a plane, train, car or boat can manoeuvre all on it's own. You can even program in many variables so it knows what to do in any given scenario.

Throw it outside that program and the computer becomes next to useless.

   Agreed with all of your post, but the part I quoted in particular is worth highlighting.

To paraphrase Jeff Goldblum from "Jurassic Park", sometimes people spend so much time trying to see if they can do something that they don't stop to think if they should do it. Frankly, I'm not sold on it being a good idea even after a viable artificial intelligence comes into existence. It may be possible, but that doesn't make it a good idea.

JMHO...
ETOPS = Engine Turns Off, Passengers Swim
 
User avatar
flyingturtle
Posts: 5684
Joined: Mon Oct 31, 2011 1:39 pm

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 11:49 am

For single-pilot ops, we'd need surveillance equipment in case the pilot is incapacitated for any reason. F/As would need to check upon him from time to time. For another reasons, single-pilot ops are possible - in small planes, where everybody can see if he's wide awake or deep asleep.

Say goodbye to reinforced cockpit doors.


David
Reading accident reports is what calms me down
 
bueb0g
Posts: 656
Joined: Fri Jul 02, 2010 5:57 pm

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 11:53 am

Quoting spink (Reply 10):

From a technology perspective, we are actually already there. From an infrastructure perspective, we have a ways to go. The technology is certainly already there to make a pilot-less plane and has been for several year

Completely incorrect. We have nowhere near the technology, no matter the infrastructure changes, to make an automated airliner anywhere near as safe as a piloted one. You could make an automated airliner, sure. Would the accident rate be 1 in 10 million..? Not a chance. The term "pilot error" is bandied about a lot but in reality, most "pilot error" accidents happen when something goes wrong and the crew fails to react in a proper manner. In many of these instances a computer - even the most technologically advanced computer - would still crash the plane.

Many more lives have been saved by pilot action than pilot error, and computers cannot respond in an intelligent way to unforeseen events, and unforeseen events happen with some regularity in aviation.

One day, there will be pilotless airliners (or at least aircraft that fly themselves with someone monitoring the progress). That much is obvious. But it's not something anybody has to worry about for well over 60 years, maybe even 100 or more...
Roger roger, what's our vector, victor?
 
2175301
Posts: 1580
Joined: Wed May 16, 2007 11:19 am

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 1:26 pm

Quoting lightsaber (Reply 11):
Quoting 2175301 (Reply 7):
My recollection is that in 17 of those cases anti-lock brakes result in better or the same level of control with fewer significant accidents. In the other cases anti-lock brakes produce worse results and more severe accidents

Lightsaber:
Only with a top driver at peak alertness. Porsche used to argue ABS had too many faults. Until every 944 in a 944 vs. Corvette race spun off the track in the rain. It was an endurance race. The race car driver was better than ABS until fatigue set in. Now put grandma behind the wheel and ABS is superior every time.

The same is true of aircraft. An alert pilot is superior to the automation. No doubt about that. But the automation can reduce the pilot's workload so significantly that they can then do their job.

Actually, the antilock brake testing was done with professional drivers. Perhaps not the best of the best (as the top notch Porsche drivers) - but much better than the average driver.

The fact is that most airline pilots are just average. Only a very few are the best of the best. Only a very few of them will be better than the automatics in most situations. The average pilot will not do as well as the automatics in many cases - just as the average driver does not do so well with automotive brakes.

In the end it will come down to a demonstration that the automatics have lower crash rates than pilots do (just like vehicles with anti-lock brakes have lower crash rates - even though they do worse in certain situations). That may well happen in cargo world first; and it may not happen for another few decades. But once it happens.... Pilots will then be another job displaced by technology and all new planes will be designed to be pilot-less (with perhaps backup controls available)

Have a great day,

Perry
 
planemaker
Posts: 5411
Joined: Mon Aug 25, 2003 12:53 pm

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 1:55 pm

Quoting bueb0g (Reply 28):
One day, there will be pilotless airliners (or at least aircraft that fly themselves with someone monitoring the progress). That much is obvious. But it's not something anybody has to worry about for well over 60 years, maybe even 100 or more...

Full AGI could be achieved in as little as 15 years (but no later than mid-century), and pilot-less aircraft require significantly less capability than full AGI. Your timeline of "well over 60 years" (let alone 100 years or more) is completely out of sync with technology.
Nationalism is an infantile disease. It is the measles of mankind. - A. Einstein
 
User avatar
Aaron747
Posts: 9924
Joined: Thu Aug 07, 2003 2:07 am

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 2:09 pm

Technology always has incredible potential, but the fact remains that even automation is designed and programmed by humans. There will always be potential situations where human teamwork yields results that cannot be achieved via automation.

I would cite this as an example:

http://aviation-safety.net/database/record.php?id=19930331-0
If you need someone to blame / throw a rock in the air / you'll hit someone guilty
 
hivue
Posts: 1957
Joined: Tue Feb 26, 2013 2:26 am

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 2:27 pm

I see a collision coming between increasing automation and decreasing routine, minute-to-minute flight responsibilities for the crew. In the AF447 investigation the BEA was careful to note the "startle effect" when the A/P and A/Thr packed it in and dumped the airplane in the laps of the pilots. My suspicion is that if the crew had been busily engaged in operating the aircraft and not just monitoring then they would not have been nearly so "startled." There likely will come a point (maybe not for 20 or 30 years) when the flight crew has so little to do that they become a hazard to safe flight operations.
"You're sitting. In a chair. In the SKY!!" ~ Louis C.K.
 
planemaker
Posts: 5411
Joined: Mon Aug 25, 2003 12:53 pm

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 4:25 pm

Quoting aaron747 (Reply 31):
Technology always has incredible potential, but the fact remains that even automation is designed and programmed by humans. There will always be potential situations where human teamwork yields results that cannot be achieved via automation.

In fact, automation can respond at speeds that no human can... and can fly severely damaged aircraft that no human can - probably the most :dramatic" was the test F-18 that had part of the wing "blown off" and the fight control software was able to adapt in real-time to stabilize the aircraft so that the pilot didn't even notice. Furthermore, Lightsaber has provided a personal example that he witness during flight test.
Nationalism is an infantile disease. It is the measles of mankind. - A. Einstein
 
User avatar
7BOEING7
Posts: 3039
Joined: Fri Oct 12, 2012 5:28 pm

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 10:53 pm

We are already on the path and within 10-15 years we'll see cargo ops like UPS and FedEx transitioning to single pilot airplanes with backup on the ground -- single pilot commercial ops will then follow. It's only a matter of time until airports world wide have autoland capability and the question asked after the crash is "Why was the pilot hand flying the airplane?". Look at the autopilot/flight control capabilities 30 years ago compared to today and imagine what the capabilities will be 20 to 30 years from now -- done deal!!
 
AngMoh
Posts: 981
Joined: Fri Nov 04, 2011 5:03 am

RE: Planes Can Fly Themselves. BS!

Tue May 28, 2013 11:40 pm

Quoting 7BOEING7 (Reply 34):
We are already on the path and within 10-15 years we'll see cargo ops like UPS and FedEx transitioning to single pilot airplanes with backup on the ground -- single pilot commercial ops will then follow. It's only a matter of time until airports world wide have autoland capability and the question asked after the crash is "Why was the pilot hand flying the airplane?". Look at the autopilot/flight control capabilities 30 years ago compared to today and imagine what the capabilities will be 20 to 30 years from now -- done deal!!

  

It is going to happen and it is a matter of when rather than if. Even today, while there are a cases where the pilots saved the day, there are also cases where the situation was saved by the automated systems despite of the pilots and there are cases where the pilots are the main cause of an accident (e.g. recent Lion Air crash in Bali, AF447, Afriqiyah Airways Flight 771).

When I stepped the first time on a driverless train in 2003, it was a novelty and quite a bit of PR went on to ensure the public it was safe. All new trains are now driverless, nobody blinks an eyelid or even thinks about it and there are some calls to make the remaining trains with driver driverless, because they are more smooth and stop more accurately in front of the platform doors.
727 732 733 734 735 73G 738 739 739ER 742 743 744 752 753 762 772 77E 773 77W 788 A300 A310 A319 A320 A321 A332 A333 A343 A345 A346 A359 A35K A388 DC-9 DC-10 MD11 MD81 MD82 MD87 F70 ERJ145 E170 E175 E190 E195 ATR72 Q400 CRJ200 CRJ700 CRJ900 BAE146 RJ85
 
planemaker
Posts: 5411
Joined: Mon Aug 25, 2003 12:53 pm

RE: Planes Can Fly Themselves. BS!

Wed May 29, 2013 2:17 am

Quoting 7BOEING7 (Reply 34):
with backup on the ground

Airlines through their OPS Centers and OEM's (A, B, GE, RR, eg.) are also in real-time data up- and down-link with aircraft... today. And within a few years ATC will have direct data link to aircraft as well and will be able to control the aircraft.

Quoting AngMoh (Reply 35):
When I stepped the first time on a driverless train in 2003, it was a novelty and quite a bit of PR went on to ensure the public it was safe.

Some experts are now forecasting that self-driving cars will be on the road by 2016 - and Tesla could very well be the first manufacturer as Musk is looking to incorporate the full capability.
Nationalism is an infantile disease. It is the measles of mankind. - A. Einstein
 
Mir
Posts: 19491
Joined: Mon Jan 05, 2004 3:55 am

RE: Planes Can Fly Themselves. BS!

Wed May 29, 2013 2:56 am

Quoting spink (Reply 21):
Or do you think the pilot is controlling the throttle, calculating the glide slopes, climb rates, rudders/flaps, etc during most of the flight?

I do all those things, yes.

-Mir
7 billion, one nation, imagination...it's a beautiful day
 
Passedv1
Posts: 649
Joined: Sun Oct 07, 2012 3:40 am

RE: Planes Can Fly Themselves. BS!

Wed May 29, 2013 4:19 am

Many of you on here clearly...

1. Have no idea what the role of the pilot is...hint...the actual physical flying of the airplane is not the hard part.

2. Have never actually performed an auto-land. Hint... The current auto-land system is designed to bring airliners down during low visibility conditions which almost always means fog which means little to no wind. If you have ever tried to fly an auto-land in gusty conditions or with a good crosswind you would not be so confident in your statements. There is no auto pilot system flying in an airliner today that can land the airplane to the airplanes full capabilities. The current systems solution if things get to hairy is to drop it in the laps of the pilots which happens on 1% or more of auto-lands (auto-pilot disconnect.). That would not be nearly an acceptable percentage if the computer was your only option.

3. Have never tried to get a new technology certifiedby the FAA... Good luck with that.

The top airlines of the world have a near perfect safety record that a computer would not be able to match. All of these other functions where autonymous vehicles operate have much worse safety records than aviation.

In the next 50 years...

Single pilot...probably.

Remotely piloted...possible.

Autonymous Airplanes....not for 100 years. At least 50 years after it happens for ships and trains.
 
planemaker
Posts: 5411
Joined: Mon Aug 25, 2003 12:53 pm

RE: Planes Can Fly Themselves. BS!

Wed May 29, 2013 4:56 am

Quoting passedv1 (Reply 38):
The current
Quoting passedv1 (Reply 38):
today
Quoting passedv1 (Reply 38):
The current systems

No one is talking about "today's" "current" systems/technology (most commercial aircraft have technology that is 10-20 years old - and some even older).

Quoting passedv1 (Reply 38):
Have never tried to get a new technology certifiedby the FAA... Good luck with that.

With every single new model of aircraft new technology gets certified by the FAA. Furthermore, the FAA is participating in "reduced crew" studies.
Nationalism is an infantile disease. It is the measles of mankind. - A. Einstein
 
User avatar
7BOEING7
Posts: 3039
Joined: Fri Oct 12, 2012 5:28 pm

RE: Planes Can Fly Themselves. BS!

Wed May 29, 2013 5:21 am

Quoting passedv1 (Reply 38):
1. Have no idea what the role of the pilot is...hint...the actual physical flying of the airplane is not the hard part.

True, most of the muscles you use are in your brain.

Quoting passedv1 (Reply 38):
2. Have never actually performed an auto-land.

Flown several thousand of them -- mostly good visibility, most not CAT II or III beams, head winds, tail winds, crosswinds, a couple with an engine out -- they all seemed to find the runway, although some were better than others -- don't know why you're talking, not having ever flown one. And with every new airplane they get better.

Quoting passedv1 (Reply 38):
3. Have never tried to get a new technology certified by the FAA... Good luck with that.

They're "picky" and as a passenger I wouldn't have it any other way.
 
User avatar
flyingturtle
Posts: 5684
Joined: Mon Oct 31, 2011 1:39 pm

RE: Planes Can Fly Themselves. BS!

Wed May 29, 2013 9:48 am

Well, we still have and use radio beacons.

I'm more amazed about old technologies that are still used than the new ones that trickle into our lives. We shouldn't look at our iPads and solid-state hard disks and infer that technology will progress at an equal pace in aviation.

The image of pilotless airplanes lives on. I chuckle every time I hear of a technology that is still about to change our lives, but was proposed in a 1986 popular science magazine: "In 2000, the V-22 will transform how the USMC is fighting wars!"

Or: "In 2000, we will do all our office tasks on a tiny Texas Instruments calculator!"

Change is slooooow.


And, as one a.net member said during the numerous AF447 threads: Computers are suited to boring situations where everything is happening as normal. They are always working on feedback loops. Too low? More thrust. Too high? Less thrust. Too fast? Pull nose up. Too slow? Nose down. They can't do anything they weren't built or programmed for. And artificial intelligence is still a very long way down the road. Neuronal networks on the other hand, are already used (e.g. ticket pricing), they can superbly recognize patterns. But human beings have to train them, and they can be over-trained to solve one set of problems, and thus be under-trained to solve another set of problems.

Pilots, on the other hand, are bored when everything is plain normal, as aircraft can be flown by button-pushing. But they earn their $$$ when the sh*t happens, and their judgement is needed ASAP.


David
Reading accident reports is what calms me down
 
Pihero
Posts: 4318
Joined: Mon Jan 31, 2005 5:11 am

RE: Planes Can Fly Themselves. BS!

Wed May 29, 2013 2:55 pm

This is a subject which comes back regularly on this forum, with exactly the same arguments from people - generally with an "engineer" tag to their pseudo - who dont have the first clue on what airline flying is about.
The animosity against the pilot population is quite incredible, summed by this one most asinine post I've the displeasure - but amusement - to read :

Quoting spink (Reply 15):
The reality is that airline pilots while highly trained are 99% glorified mechanical devices taking orders from ATC via radio and imputing that into flight computers. Once ATC and digitally send order to flight computers the pilots will be primarily reduced to looking for hazards, something that is primarily already done by computer.

The first question is : Can automatons fly airplanes ?
The answer is of course yes. All normal events can be programmed and the failures taken into account.
Then the first problem is : who makes the decision - on flap configuration, choice of take off speeds ?.. that too can be automated.
The second problem - still a matter of decision - comes when an abnormality occurs : the V1 principle is straightforward, but let's consider, past V1, a failure that signifies an unflyable aircraft... Could the engineer programming the automaton have considered all the factors ? Go or No Go ?
Please note that we haven't even left the ground and we are already facing a matter of life and death, which quite a few crews have solved by shear transgression of accepted SOPs.

Everything performing normally, there could be a case for automated flying.

The second problem is a matter of decision - again - once airborne.
With my limited programming experience on PCs, I understand a logic tree of IF / THEN... which is quite straightforward , even if a bit complicated... I have no quarrel with that, but as a computer programmer friend of mine acknowledges that a very safe program sees an error per 1000 lines. As only a FADEC software is made of a few 100,000 lines, no wonder we still have in flight flame outs and compressor stalls, not forgetting of course, the importance of valid sensor data inputs to the computers.
My point is I have never heard a tree comprising - and in an acceptable safe way - the term *WHAT IF ?* which introduces a totally unrelated condition to the decision-making process.
To illustrate this here is a problem that I submitted on an earlier thread :

" You are the captain of amedium-haul airliner flying from Paris to Turin. One pack is inoperative and Autoland is unavailable. Visibilities all over Europe is good and most runways are contaminated by snow / slush. During the climb, your TPIS generates a message of #3 tyre pressure 0. What is your decision ?"

I posit that if the proponents of the pilotless airliner can't answer the question, your dreams are premature, to say the least.

Quoting spink (Reply 15):
So the reality is that all the technology is there for pilot-less commercial flying.

So answer my question. Otherwise you're not talking through your mouth.

Quoting spink (Reply 21):
In the old days, there were at least 3 people in the cockpit, one just monitoring instruments: the flight engineer

When I started flying, and that's not really very long ago, they were more than 3 : you forget the navigator and the radio officer ( this one stayed until the late seventies as he was required for overflying the USSR as morse was the main en-route liaison with ground controllers and weather stations )... to reduce the role of the F/E to just monitoring the instruments denotes a total misconception of his very important role as system operator ( fuel / pressurisation and air conditioning / electrical loads...management).

There are also quite a few numbers of technically-unrelated occurrencies : How does the computer solve a medical emergency, say a diabetic coma ?
How does the computer manages a quick re-routing around a conflict area ?
Oh ! I know : someone on the ground can take over.
Problem is : a communication can be jammed / failed and software can be hacked... which implies the problem of criminal takeover of a flight. I just shudder at the thought of another 9-11 through computer hacking.

Now, let's talk about the single pilot operation.
Not a bad idea, considering that some measure of human intervention could happen.
But there is a series of problems :
1/- Do you really see one human sitting for hours alone in a two cubic meter cockpit?...and then required to quickly intervene when the situation arises ? I think you've redefined * bored to death*
2/- As flights are getting longer and longer, doesn't that pilot need a relief ? So in fact you're talking about at least two pilots on the aircraft. And if it is the case, these pilots will be restricted to 8 hours max on duty. That means that at the end of a long flight, the guy / girl in front would be at the end of his / her tiredness condition (time zones difference included)... That is really a vast improvement in safety.   

Generally self -confessed aviation fans do not bother with accident details ; proof is that when the official reports are published, there are very few who would comment on the findings, and fewer who would bother to read the whole report.

The incident of Qantas 32 "Nancy Bird" in Singapore is quite an illustration of this fact because it involved a list of decisions based on *out-of-the-box* reasonings and pure lateral thinking ; among those :
A/- an accurate assessment of the state of the aircraft : based on the damage to the flight controls - through ECAM messages-, they correctly identified a potentially life-threatening structural damage to the wing. So the decision was made about balancing the risk of the wing failure with the risk of an overweight oe-engine-out landing ... this decision process taking place when the SIC was dealing with some fifty items of abnormalk check-lists and had to be monitored, none of which could really be dismissed / ignored / waived.
Still based on the fact that they knew fuel tanks were punctured -no ECAM message this time - with the dissymetry of tank contents, they assessed the risk of fuel leaking on top of very hot brakes with a very likely fire situation which could make evacuation difficult.
B/- and this is very interesting : the computer refused a landing at Changi for a risk of overrun with the landing configuration and weight. Captain de Crespigny then tricked the computer into accepting the landing by changing the inputs to the LDPA : as it had rained over SIN, the original input was for *wet* runway, which he changed to *dry*.
It didn't change the fact that in terms of strict procedures he was taking a serious risk of overrun but he felt confident that with max braking, he would keep his aircraft within the confines of the runway : he was right... by 100 meters, at the cost of white hot brakes and four burst tyres.

Why is it that I don't believe a single second that a computer would have saved the day of Nancy Bird ?
To start with, a computer would have have information on the wing structure : that means stress gauges all over the wing spars, continuity currents...and why stop at the wing ? put some more into all stress areas of the aircraft and then build a software that would take them all into account.
Then, put another computer to second guess the main one in order to impose another solution based on external factors which it has to judge ... wre you ready for that ?

As one poster said above :

Quoting garpd (Reply 23):
This whole Pilotless or Single Pilot idea gives me the willies.
Until a fully developed true to life Artificial Intelligence is ready to go, the whole idea is a very, very bad one.

Can't agree more.

DOUBLE POST !!! MODS PLEASE REMOVE !!!

[Edited 2013-05-29 08:01:02]
Contrail designer
 
Pihero
Posts: 4318
Joined: Mon Jan 31, 2005 5:11 am

RE: Planes Can Fly Themselves. BS!

Wed May 29, 2013 2:57 pm

This is a subject which comes back regularly on this forum, with exactly the same arguments from people - generally with an "engineer" tag to their pseudo - who dont have the first clue on what airline flying is about.
The animosity against the pilot population is quite incredible, summed by this one most asinine post I've the displeasure - but amusement - to read :

Quoting spink (Reply 15):
The reality is that airline pilots while highly trained are 99% glorified mechanical devices taking orders from ATC via radio and imputing that into flight computers. Once ATC and digitally send order to flight computers the pilots will be primarily reduced to looking for hazards, something that is primarily already done by computer.

The first question is : Can automatons fly airplanes ?
The answer is of course yes. All normal events can be programmed and the failures taken into account.
Then the first problem is : who makes the decision - on flap configuration, choice of take off speeds ?.. that too can be automated.
The second problem - still a matter of decision - comes when an abnormality occurs : the V1 principle is straightforward, but let's consider, past V1, a failure that signifies an unflyable aircraft... Could the engineer programming the automaton have considered all the factors ? Go or No Go ?
Please note that we haven't even left the ground and we are already facing a matter of life and death, which quite a few crews have solved by shear transgression of accepted SOPs.

Everything performing normally, there could be a case for automated flying.

The second problem is a matter of decision - again - once airborne.
With my limited programming experience on PCs, I understand a logic tree of IF / THEN... which is quite straightforward , even if a bit complicated... I have no quarrel with that, but as a computer programmer friend of mine acknowledges that a very safe program sees an error per 1000 lines. As only a FADEC software is made of a few 100,000 lines, no wonder we still have in flight flame outs and compressor stalls, not forgetting of course, the importance of valid sensor data inputs to the computers.
My point is I have never heard a tree comprising - and in an acceptable safe way - the term *WHAT IF ?* which introduces a totally unrelated condition to the decision-making process.
To illustrate this here is a problem that I submitted on an earlier thread :

" You are the captain of amedium-haul airliner flying from Paris to Turin. One pack is inoperative and Autoland is unavailable. Visibilities all over Europe is good and most runways are contaminated by snow / slush. During the climb, your TPIS generates a message of #3 tyre pressure 0. What is your decision ?"

I posit that if the proponents of the pilotless airliner can't answer the question, your dreams are premature, to say the least.

Quoting spink (Reply 15):
So the reality is that all the technology is there for pilot-less commercial flying.

So answer my question. Otherwise you're not talking through your mouth.

Quoting spink (Reply 21):
In the old days, there were at least 3 people in the cockpit, one just monitoring instruments: the flight engineer

When I started flying, and that's not really very long ago, they were more than 3 : you forget the navigator and the radio officer ( this one stayed until the late seventies as he was required for overflying the USSR as morse was the main en-route liaison with ground controllers and weather stations )... to reduce the role of the F/E to just monitoring the instruments denotes a total misconception of his very important role as system operator ( fuel / pressurisation and air conditioning / electrical loads...management).

There are also quite a few numbers of technically-unrelated occurrencies : How does the computer solve a medical emergency, say a diabetic coma ?
How does the computer manages a quick re-routing around a conflict area ?
Oh ! I know : someone on the ground can take over.
Problem is : a communication can be jammed / failed and software can be hacked... which implies the problem of criminal takeover of a flight. I just shudder at the thought of another 9-11 through computer hacking.

Now, let's talk about the single pilot operation.
Not a bad idea, considering that some measure of human intervention could happen.
But there is a series of problems :
1/- Do you really see one human sitting for hours alone in a two cubic meter cockpit?...and then required to quickly intervene when the situation arises ? I think you've redefined * bored to death*
2/- As flights are getting longer and longer, doesn't that pilot need a relief ? So in fact you're talking about at least two pilots on the aircraft. And if it is the case, these pilots will be restricted to 8 hours max on duty. That means that at the end of a long flight, the guy / girl in front would be at the end of his / her tiredness condition (time zones difference included)... That is really a vast improvement in safety.   

Generally self -confessed aviation fans do not bother with accident details ; proof is that when the official reports are published, there are very few who would comment on the findings, and fewer who would bother to read the whole report.

The incident of Qantas 32 "Nancy Bird" in Singapore is quite an illustration of this fact because it involved a list of decisions based on *out-of-the-box* reasonings and pure lateral thinking ; among those :
A/- an accurate assessment of the state of the aircraft : based on the damage to the flight controls - through ECAM messages-, they correctly identified a potentially life-threatening structural damage to the wing. So the decision was made about balancing the risk of the wing failure with the risk of an overweight oe-engine-out landing ... this decision process taking place when the SIC was dealing with some fifty items of abnormalk check-lists and had to be monitored, none of which could really be dismissed / ignored / waived.
Still based on the fact that they knew fuel tanks were punctured -no ECAM message this time - with the dissymetry of tank contents, they assessed the risk of fuel leaking on top of very hot brakes with a very likely fire situation which could make evacuation difficult.
B/- and this is very interesting : the computer refused a landing at Changi for a risk of overrun with the landing configuration and weight. Captain de Crespigny then tricked the computer into accepting the landing by changing the inputs to the LDPA : as it had rained over SIN, the original input was for *wet* runway, which he changed to *dry*.
It didn't change the fact that in terms of strict procedures he was taking a serious risk of overrun but he felt confident that with max braking, he would keep his aircraft within the confines of the runway : he was right... by 100 meters, at the cost of white hot brakes and four burst tyres.

Why is it that I don't believe a single second that a computer would have saved the day of Nancy Bird ?
To start with, a computer would have have information on the wing structure : that means stress gauges all over the wing spars, continuity currents...and why stop at the wing ? put some more into all stress areas of the aircraft and then build a software that would take them all into account.
Then, put another computer to second guess the main one in order to impose another solution based on external factors which it has to judge ... Are you ready for that ?

As one poster said above :

Quoting garpd (Reply 23):
This whole Pilotless or Single Pilot idea gives me the willies.
Until a fully developed true to life Artificial Intelligence is ready to go, the whole idea is a very, very bad one.

Can't agree more.
Contrail designer
 
hivue
Posts: 1957
Joined: Tue Feb 26, 2013 2:26 am

RE: Planes Can Fly Themselves. BS!

Wed May 29, 2013 3:28 pm

Quoting flyingturtle (Reply 41):
And, as one a.net member said during the numerous AF447 threads: Computers are suited to boring situations where everything is happening as normal.
Quoting flyingturtle (Reply 41):
Pilots, on the other hand, are bored when everything is plain normal, as aircraft can be flown by button-pushing. But they earn their $$$ when the sh*t happens, and their judgement is needed ASAP.

I think it can be argued that the AF447 crew did not earn their $$$ (or €€€).

Certainly it is unlikely any automated system one could concieve of today could have handled Qantas 32. The problem I see coming is when there are two AF447s for every Qantas 32. Also, I think in the future there will be as much chance of having five crew routinely on the flight deck as having no crew on the flight deck.
"You're sitting. In a chair. In the SKY!!" ~ Louis C.K.
 
rcair1
Posts: 1147
Joined: Wed Oct 28, 2009 8:39 pm

RE: Planes Can Fly Themselves. BS!

Wed May 29, 2013 6:14 pm

Quoting Mir (Reply 9):
I don't think we'll ever have single-pilot airliners.

Single Pilot and no pilot are the same from a planning standpoint. You must be ready to deal with emergencies with no pilot. And remember - most crashes are the result of multiple small events, not a single one.

Quoting spink (Reply 10):
From a technology perspective, we are actually already there.

respectfully disagree.

Quoting bobloblaw (Reply 17):
Cargo would be a good test market before passengers

I'm not so sure - they are still operating in the same airspace as passenger craft. Therefore they cannot be considered in isolation.

Quoting ZKSUJ (Reply 24):
I think I've made my point

And several others as well.   

Quoting 2175301 (Reply 29):
he average pilot will not do as well as the automatics in many cases - just as the average driver does not do so well with automotive brakes.

Let's stop comparing autoland in bad weather with an engine out to anti-lock brakes. Antilock brakes are nothing in terms of the technology required. Also - they have something like 40 years of development behind them. And they fail.

Quoting AngMoh (Reply 35):
cases where the pilots are the main cause of an accident (e.g. recent Lion Air crash in Bali, AF447

In the AF447 case - the pilots were only the cause because they failed after the automatics packed it in.

Quoting planemaker (Reply 39):
No one is talking about "today's" "current" systems/technology (most commercial aircraft have technology that is 10-20 years old - and some even older)

Because it takes at least that long to put technology in place.

Quoting Pihero (Reply 42):
generally with an "engineer" tag to their pseudo

As somebody with both "engineer" and "pilot" tag - and also "firefighter" tag (which means I KNOW that the worst can happen - and does), I'm right with you Pihero. Engineers are great at tunnel vision.
rcair1
 
User avatar
DocLightning
Posts: 21606
Joined: Wed Nov 16, 2005 8:51 am

RE: Planes Can Fly Themselves. BS!

Wed May 29, 2013 6:58 pm

Quoting 2175301 (Reply 5):
The issue is not will the automated controls fail and crash a plane with people on-board. Everyone knows that will in fact happen. The question is will it happen more or less often than pilots making mistakes and crashing the plane with people on-board.

Bingo. The failure modes will be different, but it will be the frequency of those failure modes that can make a difference.

Already, the majority of recent accidents and incidents have been caused by pilot error, not mechanical or A/P error. Perhaps if an AI (not just an autopilot) had been flying AF 447...

Quoting bueb0g (Reply 28):
Many more lives have been saved by pilot action than pilot error, and computers cannot respond in an intelligent way to unforeseen events, and unforeseen events happen with some regularity in aviation.

Computers cannot YET respond to unforseen events in an intelligent way. Then again, ten years ago, the idea of a computer driving a car was absurd. I was just behind one today. It drove quite well.
-Doc Lightning-

"The sky calls to us. If we do not destroy ourselves, we will one day venture to the stars."
-Carl Sagan
 
planemaker
Posts: 5411
Joined: Mon Aug 25, 2003 12:53 pm

RE: Planes Can Fly Themselves. BS!

Wed May 29, 2013 10:43 pm

Quoting rcair1 (Reply 45):
Because it takes at least that long to put technology in place.

No it doesn't. Software is upgraded on A/C routinely and this is what it boils down to... software upgrades.
Nationalism is an infantile disease. It is the measles of mankind. - A. Einstein
 
Mir
Posts: 19491
Joined: Mon Jan 05, 2004 3:55 am

RE: Planes Can Fly Themselves. BS!

Wed May 29, 2013 11:37 pm

Quoting DocLightning (Reply 46):
Already, the majority of recent accidents and incidents have been caused by pilot error, not mechanical or A/P error. Perhaps if an AI (not just an autopilot) had been flying AF 447...

This is somewhat disingenuous. Yes there was pilot error, but there was also mechanical error as well. Pinning that one solely on the pilots isn't right.

-Mir
7 billion, one nation, imagination...it's a beautiful day
 
User avatar
DocLightning
Posts: 21606
Joined: Wed Nov 16, 2005 8:51 am

RE: Planes Can Fly Themselves. BS!

Thu May 30, 2013 12:00 am

Quoting Mir (Reply 48):
This is somewhat disingenuous. Yes there was pilot error, but there was also mechanical error as well. Pinning that one solely on the pilots isn't right.

There was a mechanical failure that was completely recoverable. While the pitot tube precipitated the incident, it was the pilot error that actually caused the impact with the water.

This is like the CFIT with the L-1011 in the everglades. The cause of the accident was not a faulty landing gear indicator light, but the fact that both crew left the flight deck and forgot to fly the aircraft.
-Doc Lightning-

"The sky calls to us. If we do not destroy ourselves, we will one day venture to the stars."
-Carl Sagan

Popular Searches On Airliners.net

Top Photos of Last:   24 Hours  •  48 Hours  •  7 Days  •  30 Days  •  180 Days  •  365 Days  •  All Time

Military Aircraft Every type from fighters to helicopters from air forces around the globe

Classic Airliners Props and jets from the good old days

Flight Decks Views from inside the cockpit

Aircraft Cabins Passenger cabin shots showing seat arrangements as well as cargo aircraft interior

Cargo Aircraft Pictures of great freighter aircraft

Government Aircraft Aircraft flying government officials

Helicopters Our large helicopter section. Both military and civil versions

Blimps / Airships Everything from the Goodyear blimp to the Zeppelin

Night Photos Beautiful shots taken while the sun is below the horizon

Accidents Accident, incident and crash related photos

Air to Air Photos taken by airborne photographers of airborne aircraft

Special Paint Schemes Aircraft painted in beautiful and original liveries

Airport Overviews Airport overviews from the air or ground

Tails and Winglets Tail and Winglet closeups with beautiful airline logos