How soon the pilotless airliner?

[shortfinals]

Flight is starting a debate on this. Can we realistically look forward to the future we dream of, or are we being written out of the equation? The pilotless airliner is no longer unthinkable - Learmount (http://www.flightglobal.com/blogs/learmount/2013/08/how-long-to-the-pilotless-airliner/)

[Ka6crpe]

I believe it is already here. From many of the threads on here it appears that there are plenty of people sitting in the front of modern jets who can't even carry out a simple visual approach, so can't really be called pilots.

[HDRW]

Dream of? Have nightmares about!

I would certainly not fly (longhaul anyway) as SLF unless there were at least two pilots at the pointy end. Anything less is just risking single-point failure.

[Flappo]

This is not gonna happen...

[fireflybob]

Decades away, if it ever happens - far too many variables.

Do I think airliners will become more automated - why yes! But there will still be "operatives" up the sharp end.

If we were to have fully automated airliners, who would be liable when the first one spears into the middle of a conurbation such as London or New York?

[grindles]

you are not alone, HDRW. me SLF want drivers at the pointy end! they ain't there, i ain't flying.:D

[The SSK]

Think it through …

A manufacturer (let’s call them ‘Airbus’ for example) has to convince themselves of the business case.
Then they have to spend zillions developing and certifying it to the satisfaction of every country it may fly to/over.
Then they have to find launch customers.
These launch customers have to promote it to their potential passengers who will almost certainly have the choice of flying on a humanly-piloted alternative. What will their selling point be?
What happens after the first pilotless airliner crash (there assuredly will be one)?

Just because it’s technically feasible doesn’t mean it will happen.

[structor]

If there is a central control room with 50 pilots and 10 flight engineers on duty to intervene with Autonomous passenger aircraft (or cargo) it had better be in the middle of Mount Cheyenne as any such centre will be the target for Al Qaeda or organised crime syndicates.

[suninmyeyes]

No doubt it is theoretically possible. Would airlines invest in the money required? Especially when they are not prepared to invest in decent communications over Africa and still use HF.

A few scenarios that have happened to me, I wonder how the remote guys would deal with it?

1) A light aircraft without transponder blundered into our path. We saw him visually and took violent evasive action. As far as I can see a remotely controlled airliner would have hit him.

2) Circumnavigating CBs. I very much doubt the controller sitting in a comfortable chair can have the same degree of self preservation and interest and understanding of the associated turbulence that the pilots at the front get. A good Captain and First Officer assess the weather using the radar and visual clues, discuss the options and then agree on the best routing to take.

3) Autopilot dropped out in cruise at FL350 and would not reengage. Aircraft flown manually to destination.

4) Medical emergency, doctor on board, immediate diversion required.

5) Outer pane of flight deck window cracked in cruise. Aircraft did not detect it. There are so many things in an aircraft that can be spotted visually but are not detected by the automatics and sensors.

Pilots can still do things that no automated airliner can do, ie look out of the window, see a runway and visually approach and land on it. I don't think there will be pilotless commercial aircraft flying around in the lifetime of anyone reading this forum. Maybe freighters.

[Yellow & Blue Baron]

Pilotless technology is there and has been around for some time.

The point is we (the pilots) and they (the passengers) do not want this.

The End.

[Ian W]

On the rare occasion that something anomalous occurs on an aeroplane, an alert sounds and all the flight and systems data for that aircraft are made available on the interface in real time, together with a systems diagnostic report. They can intervene as effectively as they could have done in the aircraft.

I am not so certain that the remote pilot would be able to pick up the required situational awareness of all emergencies where the automation has dropped out in this simplistic fashion. What if 'all the flight and system data is not there or is corrupt and the automation dropped out short finals?

Currently, ICAO does not recognize autonomous aircraft, only remotely piloted aircraft - that is continuously remotely piloted. All the current regulations for operation of Unmanned Aircraft Systems are being written on the basis of pilot(s) with full IFR ratings to fly the UAS. The beancounters would (and are) really pushing for total automation but there are many technical hurdles to overcome even before we get to the legal and then human factors ones.

It is also unlikely that commercial passenger aircraft will be flying and self separating in IFR as this would be very inefficient. The probability in both SESAR and NextGen is that aircraft will be flying precise user preferred or 'business' 4D trajectories that have been deconflicted out to 20 - 30 minutes ahead. There should be no reason for airborne separation assurance except as another layer of safety and for situational awareness. The 4D trajectories will have been deconflicted by ATC systems and negotiated with the 'aircraft' to come to an agreed conflict free 20 minute + 'contracted 4D trajectory'. This type of system has been demonstrated over 15 years ago. Both controllers and pilots moved to more 'management by exception. BUT there remained a real need for situational awareness and an understanding of what is happening all around the aircraft. Without that situational awareness I cannot see that the required Target Level of Safety could be reached.

That is not to say that in some areas these UAS capabilities might be used with pax. There are already 'optionally piloted' aircraft being used by the military. They can ferry supplies but return with casualties on board for example. Moving from that niche into general freight carrying wide-bodies or to passenger carrying is not going to be a simple step so don't worry about jobs just yet.

[fenland787]

Pilotless technology is there and has been around for some time.

The point is we (the pilots) and they (the passengers) do not want this.

The End.
Spot on! :ok:

[fireflybob]

Pilots can still do things that no automated airliner can do, ie look out of the window, see a runway and visually approach and land on it.

Well some pilots can.......

[StudentInDebt]

5) Outer pane of flight deck window cracked in cruise. Aircraft did not detect it. There are so many things in an aircraft that can be spotted visually but are not detected by the automatics and sensors. Probably a little redundant in a pilotless aircraft ;)

[VH-UFO]

I can see it happening with Cargo into maybe remote specifically designed airfields, but never with pax aircraft.

Wouldnt surprise me if someone like Fedex weren't looking at it now.

[Speed of Sound]

I don't want to put a dampener on things, but with the increase in companies such as Ryanair and the gradual de-skilling of pilots, a point will come where a pilot costs the company no more than a flight attendant. If they are going to retain flight attendants, then why spend billions (and it will be billions!) to save them the cost of two pilots?

Also you may get rid of two pilot wages by going 'fully automatic' but, initially at least, they will be replaced by the wages of all those extra maintenance, software, network managers, certifiers, system monitoring people who will be the price of the deal to allow pilotless flight.

[deltayankee]

I can see it happening with Cargo into maybe remote specifically designed airfields, but never with pax aircraft

Not just commercial cargo. I can see this also being appealing for the military, who could deliver supplies to war zones without risking crew.

[joy ride]

If I just happened to be a passenger on a pilot-less 777 whose engines died from iced fuel intakes on short finals into LHR, could I trust the Auto system to instantly put the nose down, accelerate, thus gaining enough lift and momentum to just clear the perimeter fence and put down on the grass without serious injuries?

Thought not!

It's a different matter with cargo planes and more acceptable because there would be no-one aboard at risk in a situation like that.

People on the ground beneath the freighter might see things differently!

[HDRW]

VH-UFO:
I can see it happening with Cargo into maybe remote specifically designed airfieldsThis may work in the USA but over here we're crowded enough that you'd find it hard to prang a 172 without hitting someone on the ground (OK, an exaggeration!), and there's no space to build new airfields anywhere useful.

What would be the point of taking the cargo to somewhere remote, when it's wanted at its (presumably usually urban) destination? The extra surface transport costs may well scupper the economics.

In fact I think the economics will defeat it all together - saving the cost of 2 pilots but having to build specialist airfields, with "unstoppable" electronics (what does a pilotless aircraft do when it needs navaids to land, and they're offline?) just doesn't add up. As far as I can see, the only people pushing this idea are the electronics firms who need to project future projects / revenue, and the free-thinkers who haven't considered all the ramifications. And journalists wanting to sell papers, of course.

Remember, the UPS at Birmingham AL didn't even have a glideslope (in fact neither did Asiana at SanFran) which would have made both places inaccessible to a robot aircraft.

Incidentally, we have had AutoLand for some decades, but I don't think *anyone* has AutoTaxy, even in testing? Apart from anything else, it would add to the Ground controller's workload, and I can't see them accepting that.

@joy_ride:
I agree completely, and I'd hate to be the software designer who tries to specify what to do with a double engine failure on short final. And I think everyone considering pilotless aircraft should say out loud: "Remember Sully!"

[16024]

A man and a dog. Man is there to feed the dog. Dog is there to bite the man if he tries to touch anything.

[Interested Passenger]

we haven't even managed to get main line driverless trains yet, and that only needs fairly simple speed control in one dimension. making trains obey signals is a trivial exercise, as anyone with a model railway and a modicum of electronic background will know.

so if we can't effectively control an electric motor autonomously, i think aircraft are safe for a while yet.


given how easy automatic trains should be, there must be a good reason they don't do it. public opinion.


docklands light railway or the train at Stanstead just trundle so don't count

[Speed of Sound]

with "unstoppable" electronics

Was listening to an interesting programme on Radio Four last night about GPS jamming. (apparently the increase in employees movements being tracked by GPS has led to a big increase in cheap, handheld GPS jammers.)

Anyway, according to the programme, the biggest jammers of GPS are large solar flares. Now a single aircraft losing the signal for 20-30 seconds halfway across the Atlantic is no big deal but every aircraft simultaneously losing their GPS in the skys over New York or Chicago is more of a bigger deal.

[slf4life]

Two dogs and a food/water dispenser. They keep each other company and keep SLF from doing anything stupid.

No doubt 'pilotless' ops is coming, but of course the entire aviation ecosystem would be different as well, from insurers to airlines to air traffic to public perception. Long ways off, certainly long after the tech is all available.

[rick.shaw]

Good grief - this AGAIN????

It's never going to happen - apart from in an airline CEO's wet dream. This one has been done to death. Of course it's technically possible - has been for years - decades even. But....

Anything else more worthwhile to ponder?

[tdracer]

We're not as far away from that day as some of you seem to think.

Airbus fly-by-wire flight controls can 'overrule' pilot inputs if it determines what the pilot wants to do is 'unsafe' := (Boeing FBW is somewhat less aggressive about it relative to Airbus, but also can overrule certain pilot inputs). To a large extent, on the newest airplanes the pilots are basically there in case something goes wrong with the automatics.

40 years ago, accident causes were more or less evenly split between pilot error and mechanical failure. Today, mechanical failure has faded into the background - pilot error is the predominate cause (and most of those are CFIT). Early on in the EROPS/ETOPS transition, the safety guys did a lot of work analyzing accidents where engine failure was involved. What they found was that it was rarely an engine failure that caused an accident - it was pilots doing something stupid in response to an engine failure.

Today, most automatic systems on aircraft are designed to default to the pilot when things go wrong - the rational being that the guy (or gal) driving knows more about what needs to happen in that particular situation then some programer sitting in an office years earlier. But to an increasing extent we're finding that 'being there' isn't the same thing as 'being aware' (Air France A330 comes to mind). Remember Moore's law (which should really be Moore's observation) - computing power will double ever 18 months - while humans have pretty much stagnated. Fully autonomous cars are in development, and self-driving cars have lapped racetracks faster than professional race drivers could do in the same car :eek:

We won't get there for many years, but the day will come when letting a human decide is demonstrably less safe than letting the automatics decide. When we get to that point, it would be irresponsible to let humans drive with hundreds of people in the back.

As for costs - back around 1980, when Boeing was developing the 757/767 - the FAA came out with their finding that a 3 crew flight deck was no safer than a properly designed 2 crew flight deck. The launch customers immediately came in and told Boeing to change the 757/767 flight deck to a 2 crew design. Boeing said 'OK, but it'll cost you another million dollars per airplane' (the first 7 or 8 767's were actually built with 3 crew flight decks and needed to be retrofit for 2 crew and EICAS). The airlines response was 'no problem, we'll save at least a $million per airplane each year in crew costs going to 2 crew'. That was in 1980 dollars (and flight engineers didn't make as much as pilots)

BTW, it's my understanding that the systems for full autonomous trains have been around for years - that they haven't been adopted for widespread use has more to do with the power of the unions and public perceptions than any technical limitations.

[MrMachfivepointfive]

we haven't even managed to get main line driverless trains yet

Dubai Metro is driverless. So is the one in Toulouse. There are probably more, but those I have used myself.

[Dan Winterland]

They are still going to require a pilot on the ground and a standby pilot in the aircraft for a long time after any commencement of operations. The way pilot's Ts an Cs are heading, this won't represent a cost saving. Meanwhile, the testing and certification costs will be enormous.

[EC-BRQ]

Creativity. This is the main thing. Maybe a computer can perform infinite calculations per second, but in terms of creativity, it will never be something like a human. As someone said before, think about the BA 777... this is a creative solution. The captain knew what to do without having this experience before... the computer has to be "told" about that situation to give the solution.

So... humans will be in the cockpits, if not forever, for a really long long time...

[PigeonVoyageur]

Here's an old one on the pilotless aircraft that I stumbled across on a Flying Jokes site:

Back in the 1970s, automation was creeping into many of the systems associated with large airliners. One day after the boffins and engineers had laboured mightily for many weeks, a fully-loaded Convair 880 took off from Heathrow bound for New York. The cabin crew did the normal safety demonstration and the aircraft taxied out to the active runway, lined up and took off in the usual manner. As the Convair climbed through about 26,000 feet, an announcement came from the flight-deck:-

"Ladies and Gentlemen, welcome onboard this the first fully-automated transatlantic flight from London to New York. So advanced are the automatic systems onboard this specially-equipped Convair 880, there is no actual flight crew onboard in the flight-deck, the door to which is therefore locked. The entire flight-plan, with all imaginable contingencies, has been programmed into quadruplicated flight management computers, all backup systems are duplicated and there is a fifth, entirely separate set of automatic systems in case of any unforeseen problems. So relax, sit back, enjoy the cabin service from our excellent crew, and again we hasten to assure any of you who may feel slightly apprehensive about this flight that nothing, I repeat, absolutely nothing can go wrong...go wrong...go wrong...go wrong...go wrong..."

[Agaricus bisporus]

Two dogs and a food/water dispenser. They keep each other company and keep SLF from doing anything stupid.

You have that already...

A man and a dog. Man is there to feed the dog. Dog is there to bite the man if he tries to touch anything.

This will be the way I fear. A single "minder" forbidden to intervene unless the automatics fail. We're already half way there with the children of the magenta line. It's only a step or two further down the road.

Auto taxi will be the easiest bit to solve.

[FakePilot]

If you think there can be pilotless aircraft then please do the following for me.
Mark a line on the floor. Stand exactly 1/3 of the way from either end and let us know.

[Tourist]

The technology is certainly available now, and those who believe that human eyes in the cockpit are superior to the optics in modern military aircraft are deluding themselves both in terms of acuity and in terms of all round global field of view when it comes to avoiding other aircraft/CFIT plus of course the ability to use conventional/nvg/flir combined.
The military is developing autonomous UCAVs which are going to go out, find enemy aircraft and shoot them down. This is orders of magnitude more difficult in terms of other aircraft awareness than merely replicating the TCAS/human eye configuration currently used in manned airliners to avoid hitting each other. Incidentally, I recently saw a brief that showed that in recent TCAS RAs at a large European airline, more than 50% of the RAs were executed incorrectly. A machine would not make those errors.

There are helicopters currently flying in Afghanistan carrying freight which is significantly technically more challenging than airliner automated flight, and those who don't believe there are trains all over the world operating driverless need to spend a little time on google.

RASCAL Blackhawk takes second unmanned flight « Helinews Asia ? Pacific (http://www.helinews.com.au/news/rascal-blackhawk-takes-second-unmanned-flight/)

This helicopter is a whole stage further. Whilst it is still in development, the challenges it is overcoming are a whole step change in terms of technical difficulty compared to airliner operation. It is autonomous. It can find it's own landing spots. You will notice on one of the videos that it is also able to autonomously avoid other aircraft visually.

The difficulty for automated airliners is in no way technical, it is public image, which may or may not be insurmountable.

[Ian W]

If you think there can be pilotless aircraft then please do the following for me.
Mark a line on the floor. Stand exactly 1/3 of the way from either end and let us know.

Draws line one inch long
Stands on line

and?

:8

[fireflybob]

Previous thread on this subject here - have things changed that much in 3 years?

Fully automated flight (http://www.pprune.org/questions/399954-fully-automated-flight.html)

One point re helicopters - not many airliners that can hover.....

[Hotel Mode]

Dubai Metro is driverless. So is the one in Toulouse. There are probably more, but those I have used myself.

They are Metros not mainline railways.

They have identical trains at low speeds on routes with few junctions, and, have reserved infrastructure from non automatic trains. ie few variables to account for in programming.

There are no main line automatic rail services. The worlds first automatic metro is nudging 50 years old and yet the technology has still not been applied to real main line railways.

[Ian W]

Well the military and government have been working on UAS that can carry out complex maneuvers for some time. Such as rotary wing operations from ships.

Boeing's Unmanned Little Bird lands on ship - YouTube

The old aphorism "just because you can do it doesn't mean you should" applies.

[flyer123]

....................................

[Iron Duke]

A human being has an input of self preservation .. a computer does not ....

A computer will gladly fly into a mountain if the sensors say it should ... a human would think " I do not care what the sensors are saying .. I am not flying into that mountain"

That is why there will always be pilots in the front ... maybe making less input, but there all the same ...

I.Duke

[tdracer]

There are no main line automatic rail services. The worlds first automatic metro is nudging 50 years old and yet the technology has still not been applied to real main line railways. Based on what happened in Spain last month, one could argue that we're killing people by not increasing automation on mainline railways. I'm pretty sure a computer wouldn't have tried to take a 60mph corner at 120....

As I noted previously, the slow adoption of fully autonomous rail has more to do with union featherbedding and public perception than in real technical shortcomings (with union spread miss-information contributing heavily to the public perception).

[tdracer]

A computer will gladly fly into a mountain if the sensors say it should ... a human would think " I do not care what the sensors are saying .. I am not flying into that mountain"

Yet three pilots flew a perfectly sound 777 into a seawall. Do you really think a computer would have done that?

[UPP]

I agree totally with all of what tdracer said in post #25, and in particular this bit:

'........but the day will come when letting a human decide is demonstrably less safe than letting the automatics decide. When we get to that point, it would be irresponsible to let humans drive with hundreds of people in the back.'

I'm from the avionics / computing side of things.

An example of how things evolve:

From 1970 onwards I was involved in the manufacture and testing of aviation syncros and actuators for fuel control systems on all the then - currrent military planes + Concorde (which had military-rated engines). These things cost a lot of money to make because they were lumps of metal with electrical windings, all of which had to be very accurately made to give the correct outputs. We were taking something that was intrinsically dumb and trying to calibrate it to give it the impression of intelligence by machining stuff to a few tenths of 1/1000th of an inch on a case by case basis. Go too far and it's scrap.

But these outputs were just there to give an indication of the position of a rotor relative to a starting point (how far did the pilot advance the throttles?), and sometimes how quickly that was being done (was he in a big hurry?). Very expensive to make because we're trying to get accurate info from a dumb device. And there's a chain of them, not just the one, and errors are additive.

Even 40 - odd years ago we knew that things would have to go from a 'lump of iron that's as intelligent as we can make it seem' to 'a lump of iron that only has to be able to revolve, but has sensors - as many as required for the application - but still vastly cheaper and more reliable, and then let a computer read the sensors and DECIDE what to do.

In those days, because we were involved in technology, we knew that this particular dream would eventually come about for two reasons: we knew that the then - nascent computing power was bound to get better, and also because it needed to, simply because we would get something much more accurate and cheaper at the same time. And even then, the computing power available could cope with that, but we had no devices available to put on the end of it. These days we do. But they aren't in planes.

The problem is that most companies can't go the 'how can we get the result we need, but in a totally different and better way' route. One reason for that is obvious; they are making money from how they do it now, and don't want anything to change. Also, there would have to be a large engineering change at the customers' end. Everyone is locked in. But also it's because most people don't think like that. Those who do become billionaires these days.

Clive Sinclair was a case in point of precisely how to do it wrong. He wanted to make a tiny tv. So he spent forever trying to figure out how to take a cathode ray tube (v. old tech) and bend it at right angles. He was trying to make the existing thing better, instead of inventing the LED screen.

Pilotless planes were a thing for the far future. But the future is nearer now, and it gets nearer at an exponential rate.

And that is the point. Computers are so powerful, and software has got so much better than we'd ever imagined it would by this time. Well, we didn't know what we imagined; just think back 20 years and see the difference. Then, a phone was a thing on a table somewhere, that you used to talk to people, and nothing else. Look at them now.

A pilot knows what he knows. Suitably intelligent control systems potentially know what all the best pilots ever knew, and therefore what to do.

A pilot is a single mind; these systems are / will be the products of the minds of large numbers of highly intelligent people (wish I was one).

No-one could jam the plane, they'd have to jam whatever it might be talking to. But if it's autonomous it has everything it needs built in. And a total electrical failure can be got round with multiple redundancy; when was the last time a plane had a total failure? And how would it be different if that happened to a pilotless plane?

30 years from now, well, I don't suppose I'll be around, but I'm fairly sure the tech available then would even surprise me.

To make the point, the software used on the shuttle Challenger is on record as being the most heavily - tested software ever written. It was written in the 60s, by people sitting in a room. In the next room there were people whose job it was to try to break it. They kept going around this loop until nothing the guys in the other room could do would break it. They reckon the code cost $10,000 a line to write. Sadly, when they checked the telemetry after the disaster, they discovered that the software had spotted the fuel leak and was trying to shut down the fuel flow, and thus allow the possibility of the astronauts escaping, but unfortunately the valves it was trying to send signals to already didn't exist.

So I don't think software will ever be the problem - if they spend enough getting it right - and if it performs as well as a human then there's no argument against it. The software will always know far more than a human about what a plane is doing, and far faster. It's the decision-making that has to be concentrated on. The whole thing about suddenly deciding to do something, and better than a human is, in principle, a question of hard work and testing the software to destruction time and again until it doesn't get destroyed. At that point it's at least as good as a human.

The ultimate goal is to have an autonomous system that can think and react in all situations like Sully did in one situation. In principle that's possible now.

[pull-up-terrain]

Before they start manufacturing pilotless airliners, I would love to see Airbus and Boeing manufacture a plane without serious design problems (787...cough...cough...).

Don't these pilotless aircraft require an airport with CAT III ILS when it comes to landing?

[joy ride]

What if an airport service vehicle, waiting to cross a runway after a plane had landed, had a tech or driver problem (eg heart attack) and surged out onto the runway in front of the plane?

What if one plane lost control while landing or taking off and skidded towards another on a parallel runway, perhaps a bit like Asiana 214?

Could the system make the right choices here? The pilotless heli landing is a fine achievement but it did not have to share its required space with other traffic.

Even if one day we have a system which really can cope with every possible event, there is still the problem of selling it to the public. In London we have hundreds of planes flying over us all day, if the population was told "next year it will all be "pilot-less" there would be an uproar, and it would be fueled by the more sensationalist members of the press. It could prove difficult to get people on the ground to accept it, as well as the passengers.

In this world accidents will always happen, but imagine the outcry the day the first pilot-less airliner hits buildings.

[UPP]

'What if an airport service vehicle, waiting to cross a runway after a plane had landed, had a tech or driver problem (eg heart attack) and surged out onto the runway in front of the plane?'

Simple. A human pilot might see it, and might eventually react. Fast enough to avoid an accident? Who knows.

A computer-controlled plane would spot it and would attempt to react like greased lightning. Fast enough to avoid an accident? Who knows. But it would have a better chance than a human.

As an aside, we should actually use chimps to fly planes, when it comes to visual acuity. I saw a jaw-dropping demo once.

A chimp had a grid flashed up on a screen in front of him. Some of the squares in the grid were lit up. After a few seconds they were switched off. To get his reward, the chimp had to touch all the squares that had been lit up. He did. The interval got ever shorter until, eventually, human eyes never saw them light up in the first place. But the chimp always got it right.

A suitably-equipped plane will always be faster than you could ever be in trying to avoid this sort of accident.

Remember, sensors far better than a human can be made for anything you need to know. It's the software that matters.

[egravitics]

Soyuz / ISS rendezvous has now been fully automated to give a 4-orbit ascent and docking in under 6 hours. Ops efficiencies will drive implementation of automation in air flight too resulting in piecemeal introductions - easiest and greatest savings first for commercial contexts. A similar time-line will evolve for the military with differing concepts of what constitutes a saving. Once in that future the greatest threat to air safety becomes each still-human-operated rig which might decide to not co-operate with the global airborne separation system (GASS). Following risk analysis and agreements with unions crew in the pointy-bit will be retired from service and everyone in the air will feel much safer knowing they're not going to have a couple of human-beings being human at inappropriate moments (eg, caught up in a discussion about company policy and cruising on past the airport while GASS directs everything over, under and around them).

[Mk 1]

Not just commercial cargo. I can see this also being appealing for the military, who could deliver supplies to war zones without risking crew.

deltayankee: Already being done. 2 airframes have been operating in afghanistan for at least a year IIRC.

Lockheed Martin · U.S. Marine Corps to Keep K-Max Unmanned Cargo Re-Supply Helicopter in Theater for Second Deployment Extension (http://www.lockheedmartin.com.au/us/news/press-releases/2012/july/ms2-0731-us-marine-corps-to-keep-k-max-unmanned-cargo-helicopter.html)

[parabellum]

The technology doesn't even make it into the frame, never mind a poor fourth.

The first pilotless aircraft carrying a revenue load of pax will happen;

a). When all terrorism has been completely contained and

b). When insurance underwriters agree to fully cover passenger and third party legal liabilities for such an aircraft, (which currently run to billions of dollars of cover for piloted aircraft).

c). When a major airframe maker is prepared to invest billions of dollars in R & D for such a project believing they will have, at least, a break-even market (A380 still about 200 airframes away from break-even).

The actual cost saving from removing pilots will be miniscule when compared to the cost of R & D alone, never mind all the other associated costs like setting up ground control stations etc. etc.

This subject has been flogged to death so many times!:rolleyes:

[DWS]

They reckon the code cost $10,000 a line to write. Sadly, when they checked the telemetry after the disaster, they discovered that the software had spotted the fuel leak and was trying to shut down the fuel flow, and thus allow the possibility of the astronauts escaping, but unfortunately the valves it was trying to send signals to already didn't exist.


There are no ' valves' to shut off the 2 solid propellant boosters used-

Yes the lox hydrogen engines did have shutoff valves- but when the booster leak of hot gasses penetrated the center tank - it was already too late

Curious of source of yur comment however . . .

[MG23]

Soyuz / ISS rendezvous has now been fully automated to give a 4-orbit ascent and docking in under 6 hours.

Docking with a space station is easy compared to landing an airliner, since you're moving a few feet or inches per second with no other traffic to get in the way, and has still been problematic to implement.

[MG23]

Curious of source of yur comment however . . .

The Challenger transcript shows the engine management computers attempting a clean shutdown of the shuttle main engines during the breakup as they saw fuel pressure drop. I presume that's the source.

Edit: it's worth noting that we almost lost Challenger a few flights earlier due to defective engine sensors which caused the computer to shut down one engine during launch and wanted to shut down a second. The software appears to be easier than making sensors that work reliably, as we also saw in the AF447 case.

[Speed of Sound]

The airlines response was 'no problem, we'll save at least a $million per airplane each year in crew costs going to 2 crew'.

Are flight engineers really paid that much? :confused:

[egravitics]

Docking with a space station is like drawing alongside a jetty - you're moving slowly relative to it and you want to stop in a pre-determined place. Unlike parking next to a jetty you have to crash into a space-station which unlike a jetty is moving too. So it's important to manage the energy carried by the docking craft as there's no place for any excess to go except into changing the trajectory and/or attitude of the combined crafts. The automated docking sequence has no concept of self-preservation only parametric boundaries linked to sensor inputs plus any optional input from the PAX turning into crew. The ascent to the ISS orbit and chase is more equivalent to the landing phase for an aircraft - find the destination, alter altitude to reach it, manage energy and trajectory while not running out of fuel, making a bad decision or flipping the wrong switch. Prior to automation this phase took 2 days, now its 6 hours. In an article introducing the new regime, SpaceflightNow (http://spaceflightnow.com/station/exp35/130305rendezvous/) observed: "The fast-track rendezvous reduces the workforce required to operate the Soyuz spacecraft. Instead of staffing the Soyuz control center in Russia for more than two days, a full complement Soyuz controllers will only be needed for a day". While taxpayers are hardly likely to be rejoicing with the news of this saving, it does mean that the productivity of the industrial process of spaceflight is increased - more flights from the same cost of ground facilities and staff, from the same level of investment. Of course that is another way of interpreting the savings made by automation with respect to flight crew - the efficient deployment of capital within businesses flying freight or people or productivity gains for tax-payers in government run outfits.

[FakePilot]

Draws line one inch long
Stands on line

and?

See, you can do it but the computer can't.

[Tourist]

parabellum.

I believe this covers point c)

http://www.baesystems.com/magazine/BAES_051920?_afrLoop=62799178822000&_afrWindowMode=0&_afrWindowId=null&baeSessionId=TnTDSXYTxnT2TDDrvRJHWvntvPlCYvvSFMvyv0FDWPvt6pY M6kvz!2091114453#%40%3F_afrWindowId%3Dnull%26baeSessionId%3D TnTDSXYTxnT2TDDrvRJHWvntvPlCYvvSFMvyv0FDWPvt6pYM6kvz%2521209 1114453%26_afrLoop%3D62799178822000%26_afrWindowMode%3D0%26_ adf.ctrl-state%3D13oqag6xku_4

[joy ride]

How about a scenario in which a mis-calculation leads an airliner to run out of fuel. The flight crew spot a disused airfield and manage to glide in safely, without harming any one on board or any of the public at a meeting on the runway. A disused airfield would not have the equipment for a pilot-less landing.

Or a scenario where an engine failure takes out the hydraulic lines and all flight controls. The flight crew gradually realise that they can keep partial control of direction and altitude by altering the thrust of one engine. They manage to make it to an airport and despite a crash landing there are significant numbers of survivors.

Of course, I am aware that sometimes flight crews have made th wrong decisions and caused an avoidable accident. All humans are fallible.

In my opinion human fallibility extends to those who design automatic systems and have total faith in the technology and total faith that they have taken all eventualities into account.

If/when a pilot-less plane does crash, and regardless of what the aviation industry thinks, the public outcry and insurance questions could be insurmountable.

[mross]

In the link in post #1, Learmount says that the remote pilot will have access to data "in real time" Has he ever made a satellite phone call? The latency is about one second and this makes even phone calls very difficult for the uninitiated. Trying to remotely fly a plane with a one second delay in inputs and feedback will be almost impossible. The plane has to be autonomous in regards to flight surfaces etc. Only the routing and strategy could be controlled remotely.

[Tourist]

Joyride

If you watch the videos in my previous post you will see that autonomous vehicles are more than capable of landing at a disused airfield without any landing aids, in fact they have many advantages over a piloted aircraft in that situation, ie they can do it IMC/in the dark and unlike a pilot who has never practised for real can be programmed with the full glide performance etc

re your point about hydraulic failures etc.

There have I believe been three big occurrences. Sioux City, DHL Iraq and Korea?
The first went not too bad, the second went very well, and third was a total disaster.


Propulsion Controlled Aircraft (PCA) | NASA (http://www.nasa.gov/centers/dryden/history/pastprojects/PCA/index.html)

The work has already been done to solve the problem for computers.

I reiterate.
The problems are in no way technical.

[fireflybob]

Reference the Jetstream flight above here is a quote from a previous thread:-

They are not testing auto take-off or auto-land - the two pilots on board will handle those phases of the flight

Full thread here:-

Testflight with 'unmanned' Jetstream (http://www.pprune.org/rumours-news/501538-testflight-unmanned-jetstream.html)

Maybe not so "automated" as it's trumped up to be?

[iceman50]

If you watch the videos in my previous post you will see that autonomous vehicles are more than capable of landing at a disused airfield without any landing aids, in fact they have many advantages over a piloted aircraft in that situation, ie they can do it IMC/in the dark and unlike a pilot who has never practised for real can be programmed with the full glide performance etc


Where does this "rubbish" come from, there are way more variables than you can think of than "can be programmed with full glide performance". How is this "autonomous" vehicle going to "decide" to land on the disused airfield in the first place?

[rcsa]

Thanks UPP - that one post made the entire thread worth reading.

[mitrosft]

Long ago ( 1988 ) the russians who make computers out of tank spare parts flew space shuttle Buran pilotless, up to space and back.

Now you guys say its not feasible ? I bet the autopilot of Buran was less powerful than iPhone 3.

It's not a matter of technology now, it's the matter of legislation. The tech is there already.

[fenland787]

sometimes flight crews have made th wrong decisions and caused an avoidable accident. All humans are fallible.

In my opinion human fallibility extends to those who design automatic systems and have total faith in the technology and total faith that they have taken all eventualities into accountExactly, which is why the current arrangement - using a combination of the two - is a great solution to flying large tubes full of people around a crowded sky?

[Tourist]

iceman

Did you bother to watch the videos?

That fully autonomous helicopter is able to scan the terrain in real time and make an assessment of rough unknown terrain and decide where to land.


Throwing an airliner at a disused runway is several orders of magnitude easier than that. It is extremely easy for a modern autopilot to outperform a human pilot in all stages of flight, and has been for decades.

ILS Cat3B is stoneage

The F35 is going to be able to land on a moving carrier automatically IMC in the dark in a storm.

The tech challenges have already been achieved.

fenland787

Have you considered that the pace of technology is very impressive and automation keeps getting better.
Could you honestly say that the quality of todays pilots is as good as those of yesteryear and that tomorrows pilots will be as good as todays?

As tech improves and pilots get ever closer to an automation observer, totally unable to add anything useful to the mix, this can only end in one place.

The one area that the human in the mix is better currently than a computer is in dealing with totally new situations, and this advantage is being eroded very fast by learning computers.

Computers have so many advantages.

Fireflybob

Have you wondered why they didn't bother testing auto takeoff and autoland?

Perhaps because those phases are already in use in operational aircraft including in the case of autoland just about every airliner on the planet?!

[framer]

I'm not convinced that technology isn't a problem.
When the engine flew to bits on the Qantas A380 in Singapore there were literally hundreds of prioritised warning messages generated. Most systems had been degraded in some way or another. Each time the pilots cleared a warning message the next one appeared below it. The pilots ultimately ignored many of the alerts and started concentrating not on what they had lost but on what they had that worked and landed the aircraft. Once on the ground an engine was unable to be shut down as the physical damage to the controlling system was too great. Eventually all pax deplaned via stairs and no injuries occurred.
What would have been the outcome if this was a fully autonomous aircraft ? If the aircraft did make it back to ground would it have had success in keeping the pax onboard and away from the fuel, hot brakes and running engine? Would it have even considered doing so?

[framer]

Throwing an airliner at a disused runway is several orders of magnitude easier than that. It is extremely easy for a modern autopilot to outperform a human pilot in all stages of flight, and has been for decades.The A/P in the new 737-NG can't even begin a descent without exceeding MMO when you've entered accurate winds in high tailwind conditions. It also gives up and disengages on approach in turbulent conditions sometimes.there are also many other situations when it hands the aircraft back to you.
The radar in it paints green when the tops of some nasty CB's are full of ice and often paints magenta below 20,000ft in smooth stratus. If the air/ground switch gets dirt in it the aircraft does all sorts of strange things like opening outflow valves and disengaging automatics.
Human error that is removed from the flight deck will be replaced by human error from and Engineer or Software programmer or factory worker fabricating parts.

[framer]

Would these aircraft weigh themselves? I got sent a zero fuel weight from our load control department that was intended for another aircraft a few weeks back. It was ten tonnes lighter than we were. Plenty of human error going on in the department responsible for sending weight and balance information to aircraft. We will surely have to automate the data entry for uplinks as well, but hang on, the chap weighing the freight and bags could put a decimal point in the wrong place..... we need to develop a robot for weighing bags.....surely we can remove this pesky human error.

[fireflybob]

Tourist, so according to you it's all done and dusted - I bow to your superior knowledge.

So you are saying automation can cope with all possible scenarios?

What's your answer to framer's post re the Qantas A380 failure? Would your magic systems cope with such a failure?

You discount Cat IIIb as outdated which I agree with but what happens when the GPS goes down?

[fireflybob]

Tourist, please list aircraft types that have automatic take off - am assuming auto takeoff can cope with an engine failure after V1, dump fuel and return for an engine out approach followed by auto Go Around and Autoland with one engine inoperative?

[Ian W]

See, you can do it but the computer can't.

What you mean is that the software designer did not foresee the problem. Computers do just what they are told by the software written by humans.

[Ian W]

Before they start manufacturing pilotless airliners, I would love to see Airbus and Boeing manufacture a plane without serious design problems (787...cough...cough...).

Don't these pilotless aircraft require an airport with CAT III ILS when it comes to landing?

No aircraft needs ILS to land in poor visibility manned or unmanned. GLS and variants of it are capable of landing anywhere with the same accuracy as IlS and without the need to have a long straight in approach and no multipath risks. A GLS approach can be a continous curve or multiple curves. Like the RNP approach to Washington Reagan.

Give the UAS a terrain map and it will be able to navigate 'visually' using the shape of the terrain.

All these systems are tested and available now.

[UPP]

@DWS

Oooops, I put in a zero too many. It was $1000 per line. Sorry about that!

http://history.nasa.gov/sts1/pages/computer.html

[FullWings]

The day I fly to another country in a pilot-less airliner, I will have been gently woken up in the morning by the house, which will have prepared my breakfast and laid out appropriate clothes. The journey to the airport (if such a thing still exists) by autonomous flying car will have been pleasant and allowed me to catch up on the latest events and my personal and business (if such a thing still exists) life.

Checking in, luggage and transport to the aircraft is all taken care of by my software personal assistant, which also looks after the arrangements at the far end, like further transport and immigration (if such a thing still exists).

None of the above are technically infeasible in the future, let alone now. I think we do have a habit of totally underestimating the time to mass adoption of things that are possible but not yet part of our cultures. Pretty much all business today could be done remotely through conferencing, telepresence, etc. but the demand for work-related travel and the need for face-to-face contact is still very much there. We are human, after all. Barring some kind of Singularity (http://en.wikipedia.org/wiki/Technological_singularity), pilot-free passenger A3XXs and the like are not going to be operating overhead for some considerable time, probably not during the careers of any commercial pilots who are qualified now...

[Tourist]

Nope, I am not saying that automation can cope with all possible scenarios.


Are you saying that manned cockpits can cope with all possible scenarios?
The inability of an Air France crew to fly an Airbus and an Asiana crew to land in CAVOK suggest otherwise.

What I am saying is that the crossover point between the safest operator of an aircraft being a person or a computer is either very nearly here or very actually here.

People seem to want to hold an automated aircraft to a higher standard of safety than a manned one. The public may agree, but only to a point I feel.


The public has quietly accepted many tech advances over the years that people said "they will never accept that!"

1. Cars that go faster than a man with a flag.
2. Trains without drivers.

to name just two.

Those on here that keep taking about 787 380 etc being unable to cope miss the point entirely.

They were never intended to cope, so they can't.

It is not technically difficult with todays level of technology.

GPS is just one of the many many ways an aircraft could fly an approach.

FLIR/ground mapping radar/INS/GPS can all be seamlessly combined without any new leaps of tech or risk.

...and yes, an aircraft can weigh itself.

[fenland787]

Have you considered that the pace of technology is very impressive and automation keeps getting better.Yup, I've been automating things with computers since the 1970s

Could you honestly say that the quality of todays pilots is as good as those of yesteryear and that tomorrows pilots will be as good as todays?Not qualified to judge but strongly suspect they are, just the skills needed have changed a bit
The one area that the human in the mix is better currently than a computer is in dealing with totally new situations, Agreed
and this advantage is being eroded very fast by learning computersNot yet really, 'baby-steps' stage at the moment, and by the way, does the phrase: "I'm sorry Dave, I'm afraid I can't do that" ring any bells?

Computers have so many advantages. and disadvantages too!

All of which brings me right back to my post:

"The current arrangement - using a combination of the two - is a great solution to flying large tubes full of people around a crowded sky"

[Clandestino]

PPRuNe, thank you for this thread. At least now I see that it's not only aviation infested with folks spurting out long diatribes demonstrating they are pretty clueless about what they claim to be experts in, computer science is affected too.

Not to mention that mix of aeronautical and computers ignorance is quite amusing to read - especially when combined with unjustified assertiveness.

[MikeBanahan]

There's a lot of people saying here 'can it cope with this, can it cope with that?' when the answer will probably be no - not in a carefully selected scenario where automation is obviously at a disadvantage.

What will make the difference is not all those outliers where occasionally a superhuman effort pulled a rabbit out of a hat in remarkable circumstances. It will be the routine elimination of fatigue, inattention and other human related CFIT and similar accidents.

When aircraft are automated there will be accidents and loss of life. When humans fly, there are accidents and loss of life. Those scenarios and events will be obviously different from each other and there will be a clear distinction between human-engendered incidents and those that are down to automation.

As automation improves, the balance of probabilities will shift. At some point, it will clearly be evident that machine-flying, overall, would lose fewer lives per year than hand-flying, even though the accident profiles are different. Some number of years after that, well after it becomes obvious to anyone who can read the statistics, the regulatory bodies and the industry will adapt.

If it's clearly proven that the odds of surviving a flight are better when it's automated than hand-flown, I'll be buying tickets and so will the public.

[MG23]

Long ago ( 1988 ) the russians who make computers out of tank spare parts flew space shuttle Buran pilotless, up to space and back.

Now you guys say its not feasible ? I bet the autopilot of Buran was less powerful than iPhone 3.

Yes.

Buran took off through its own airspace in good weather, flying under power for about eight minutes. It flew back through its own airspace for about thirty minutes and landed in good weather.

That's trivial compared to taking off and landing through congested airspace in poor weather, like an airliner flying in and out of Heathrow.

The American space shuttle was capable of the same, with some wiring mods. But if Columbia had tried it, we'd have lost it on the first flight, because the guidance equations programmed into the computer were incorrect; the crew saw the computer wasn't flying properly and took over for part of the re-entry.

The world is full of software projects where they hack together something very simple and do it so fast that everyone expects the full system with all the bells and whistles to be available next week. But they're still trying to get it to work in the real world years later, because it's all the unexpected cases that kill you.

[Tourist]

A Swarm of Nano Quadrotors - YouTube

Aggressive Maneuvers for Autonomous Quadrotor Flight - YouTube

Aggressive Quadrotor Part II - YouTube

Aggressive Quadrotors Part III - YouTube

Tell me again how computers can't fly better than people?

These are mere cheap toys played with by students with tiny budgets.

They cost less than a display screen in an airbus yet autonomously operate amongst their fellows.

The improvements in air traffic safety and fuel economy are obvious when this sort of system is in place.

[EC-BRQ]

Why do we have to take out of the plane the creative device (also called "Pilot"), when we have other devices that try to eliminate the human error but those are not so creative? Is it not better to maintain both, to try to eliminate dangerous "interferences" and to use the better of the two?

All of us know what a beancounter would say but... under a fully technical point of view... why do we have to rely in only one "system"?

[Dryce]

I think that the question could be alternatively phrased:

How soon the co-pilotless airliner?

How soon the remotely piloted airliner?

[LLuCCiFeR]

Thanks to Pay to Fly we already more or less have 'pilotless' aircraft, at least for accounting purposes.

Additional benefit is that there are still two very convenient and cheap scapegoats sitting up front for when the proverbial sh!t hits the fan... :suspect:

Just call it "pilot error," blame the crew and the everyone (read: responsible authorities, postholder managers, pilot schools selling P2F and P4T scams) can continue business as usual...

Why would any, any aircraft builder consider squandering the skyhigh development costs and assume the enormous liability for a truly pilotless aircraft, when you can put two button pushing MPL slaves up front who even pay for the privilege of following a magenta line?

Yeah, go ahead and call me a cynic, just project the developments from the last 10 years forward another 10 years... :ouch:

[MG23]

Tell me again how computers can't fly better than people?

These are mere cheap toys played with by students with tiny budgets.

I refer the honourable gentleman to the last paragraph of my previous answer.

[Dimitris]

MG23,

What your previous answer says is that ATC will still be relevant in a pilotless plane. It will also solve the 'using standard aviation English in RT' problem. No more problems with accents, bad english etc etc. There will be other problems for sure.
Pilots that use as argument the 'navigating in congested environment' in order to keep humans in the cockpit, just make a case for the opposite.

And by the way, autonomous spaceflight and rendezvous (Bouran, ATVs et.al.) are a lot more complicated that you make it sound (i.e. great engineering achievements) but probably autonomous airflight is now more advanced than autonomous spaceflight.

[MG23]

What your previous answer says is that ATC will still be relevant in a pilotless plane.

I don't see where you read that into the paragraph I mentioned.

What I was pointing out was that the last 10% of a complex software project takes 90% of the time, or more when that 10% will kill people when it screws up. Yet people regularly see a project that works well in perfect conditions in its early stages, and claim they can have it operational very soon.

Writing software that works in perfect conditions is easy. Writing software that doesn't crash and burn when something unexpected happens is hard. Writing software which has to make decisions like 'I can continue this landing approach and maybe get down with everyone alive before the remaining engine fails, but if it fails early I might crash into that housing estate and kill everyone on board and hundreds of people on the ground, so do I risk it?' is pretty much impossible.

To solve that, any 'pilotless' airliner is going to end up requiring some kind of remote piloting capability, which just creates another way for bad guys to make it crash and burn.

[Tourist]

Vijay Kumar: Robots that fly ... and cooperate - YouTube

re atc required.

Watch this video all the way through.

Some of it is not particularly relevant though still quite interesting, and some of it is quite simply the future of atc.

Autonomous with no central control.

[MG23]

Another issue to consider: software upgrades.

Suppose there's an AF447-style crash, this time with a pilotless airliner. Multiple simultaneous sensor failures trigger a software bug ('we only tested dual failures, because a triple failure should never happen'), and the plane flies into the sea.

Bug has to be found, patched and tested. Now hundreds of airliners have to be upgraded. Each upgrade probably requires taking it out of service for a couple of days to install and test before you can risk putting passengers on board.

Or the airline can keep it flying. It's a rare condition, not likely to happen again before the next scheduled maintenance. But it might, and then everyone dies.

Who's going to take the risk of not upgrading? Who's going to fly as a passenger if they don't, knowing there's a fatal bug in the software flying the plane and there's nothing anyone can do to save them if it's triggered?

These are the kind of real-world issues that kill the grand utopian plans.

[framer]

Those scenarios and events will be obviously different from each other and there will be a clear distinction between human-engendered incidents and those that are down to automation.
There will be no clear distinction at all. If there is an automation event, do you not think that is also a human-engendered
Like I said in a previous post , the human error is simple moved further down the chain. You can never get away from it.
One assumption that non- pilots make a lot in this thread is that pilots cause more problems than they create. The opposite is true. From planning stage to disembarkation we are recognising and mitigating errors made by software designers ( incorrect coding of waypoints etc) engineers, ATC, aircraft designers, meteorologists, baggage handlers, load control staff etc. Notice any similarities between these parties? They all have human input to the system, sometimes years in advance.How do we remove that human input? We can't, we would just be moving it away from the aircraft.
Pilots, like every other human make mistakes, but don't forget that we also recognise and fix them.

[AndyPandy068]

While it might be the bean counters dream, I can't see me or many others getting into an autonomous aircraft with no pilots anytime soon.

[neville_nobody]

Even if it did fly I would hate to see the insurance premiums.

It would be much easier to hijack not to mention the risk of 'black swan' events where a human would save the day. QF 32, UA 232, BA 38, DHL Bagdad, etc aren't going to be saved by a UAV.

[Uplinker]

Answer to the OP:

Never.

U

[parabellum]

As I said in an earlier post and now repeat/expand;

First you have to find a manufacturer who is prepared to spend billions on R & D and currently that doesn't happen until the likely market has been thoroughly canvassed and letters of intent signed sufficient for the plane maker to think it is, in their opinion, viable.

Second you have to convince the insurance industry that it is a worthwhile risk and that a bunch of computers won't cause a major crash in the centre of a big city and leave them staring at a bill for, literally, billions. We human pilots have managed to convince them we are worth the risk but until the computer industry can do the same, with the aid of an airframe maker, the insurance industry won't touch it.

I don't see these two obstacles being cleared any time soon, there simply isn't enough money available.
It will prove cheaper to continue with human pilots, albeit backed up by ever improving automation. (Which may produce even cheaper pilots!).

Finally, the entire system worldwide will have to be terrorist proof.

Tourist - I appreciate BAe Systems have done a lot of work in the automation field but that doesn't come close to the cost of building a full size passenger transport aircraft and completing a full flight test programme right through to certification.

And to the poster who said he would buy a ticket and so would the public, methinks you must be related to Methuselah.;)

[tdracer]

Curious if you guys are familiar with the FAA "Next Gen" air traffic control system in development? You know, the one that will basically take the human element out of the air traffic system and let computers run it, because the humans are too fallible? :rolleyes: Just asking...:E

Listen, I deal with flight critical software on a daily basis. And I've heard all the complaints before - except they were directed at FADEC engine controls, glass flight deck displays, and fly by wire flight controls - systems that are now commonplace. I remember, back when FADEC engines were being introduced, there were long-timers that were insisting we had to do a full functional test engine run every time we replaced a component because "the software might be loaded wrong". :sad: I also remember so-called experts that insisted that before we could certify FADEC or FBW, we'd need different software in each control channel so a single s/w error couldn't take down the airplane (no, we don't currently do that, although some of the early systems did).

In theory, a single software error in a FADEC could cause all engines to loose 90% thrust as soon as the aircraft leaves the ground. I doubt the best stick and rudder pilot in the world is going to be able to turn that into a happy ending (at least not without a huge helping of good luck). Similarly a single error in the FBW code could command full nose down at 200 ft. during final - regardless of the control input. Pretty much a guaranteed really bad day. Any one out there refusing to fly on FADEC or FBW airplanes because the s/w can overrule what the pilot commands?

Computer power is growing exponentially - assuming Moore's law holds for another 10 years, computers in 2023 will have ~100 times the processing power that today's computers have. The automatics in the 777 - which most consider to be pretty sophisticated - are for the most part 20 years old. That's pretty much antique in the electronics world (in fact a major problem is component obsolescence - we simply cannot get the electronic bits that we certified with 20 years ago, and certifying even drop-in replacement parts can be slow and hugely expensive).

By necessity, electronics advances in commercial aviation are slow and tedious. It used to be we copied military - only a few years later - but military developments have slowed (how long has the F-35 been in development?), and make up an insignificant portion of the electronic market. Used to be we used Mil Spec electronics for flight critical avionics, but to a large extent they are not even available anymore - we're using Commercial Off The Shelf stuff, just screened to higher standards that what's in your laptop.

Watch what's happening in the automotive world, where product cycles are 4 or 5 years (many consider the 777 to be 'new' - cert and EIS was in 1995 - compared a 1995 car to a new car lately?). Experts are saying that fully autonomous cars will become common in 10 to 15 years. At first, it'll be an option on the high end luxury cars - you know, something to take you to the office while you do paperwork, or get you home from the pub after you've had a few too many. But soon they'll spread down-market (see anti-lock brakes, stability control, etc.). Then you'll be able to use an app on your phone (or whatever it's morphed to then) to summon a driverless taxi that'll take you to your destination and automatically charge your credit card or bank account (why did I just visualize Johnny Cab in Total Recall ;))

Before long, it'll become apparent that the accident rate for autonomous cars is a small fraction of those with drivers, and the accidents that do occur are nearly always the fault of a human driver. People who insist on driving themselves will be banished to back roads and such and hit with huge insurance premiums.

Will this happen soon? No, probably not in my lifetime, at least I hope not - I like driving (fast) :) (and I hope I'm good for at least another 30 or 40 years:D). But I have little doubt it will happen. Imagine a future where autonomous cars have made traffic accidents a rare occurrence (and fatalities almost non-existent). A future where the drive to the airport is no longer the most dangerous part of air travel because the human pilots still cause CFIT accidents on a regular basis.:ugh:

Do you really think, in that future, people will continue to insist that humans pilot the airplanes they fly on?

[stilton]

30 years (not saying it's right)


:(

[mitrosft]

Yes.

Buran took off through its own airspace in good weather, flying under power for about eight minutes. It flew back through its own airspace for about thirty minutes and landed in good weather.

That's trivial compared to taking off and landing through congested airspace in poor weather, like an airliner flying in and out of Heathrow.

Agreed.

But my point was that computing power as of today allows it. The sofware may still need some testing ( years maybe) BUT ! Remember the Bodenzee crash, TCAS did know exactly were both aircraft were at 50 miles before impact, it was dumb Russians with poor English command and same sllepy Swiss who did it.

Now TCAS with some added soft can navigate you in Heathrow air space just as well and faster, because it will start maneuvering the plane BEFORE a human ATC can pronounce GOLF CHARLIE bla bla bla and pilot will acknowledge it.

I am not saying that Im fond of automated planes. But having graduated from Aviation Institute with engineering degree in automated systems I vote for automated cars every day. This will bring down road deaths from more that 1000 per day globally to maybe 10 just because the Intel chip cannot drink ;)

[joy ride]

I am very impressed with what modern technology can do, but it is not just a question of how reliable the system is, it is a question of how humans perceive it. Governments will have to deal with uproar from the press and public and possibly insurers when pilot-less plane flights are announced. People have long memories of unsinkable ships sinking and "perfect" systems failing.

Once this is dealt with satisfactorily the airlines and authorities will then have to decide if there will be no pilot at all, or perhaps one "just in case".

How many times have drunk/rowdy/aggressive/disturbed passengers refused to be controlled by flight attendants, yet when the Captain comes out and has a few firm words they HAVE finally calmed down?

In the Quantas A380 incident mentioned earlier the entire crew had to keep the passengers as calm as possible for an hour before letting them disembark safely. I have little doubt that the presence and explanations of the HUMAN pilots would have been critical in avoiding a riot and break out.

Would the passengers have responded favourably to a synthesised voice telling them to remain in their seats when they can see fire outside? After that serious tech failure, would they trust the e-voice?

So perhaps we will have to keep a single pilot in our "pilot-less" airliner, able to take control in an extreme situation.

So this pilot enters, performs all necessary checks and tests, then sits back while the plane gets on with the journey. Four hours later there is a major problem and suddenly there is a real need for a fully alert, fast-thinking pilot, but our solitary bored hero has fallen into a state of total torpor. Perhaps our pilot will have to be fitted with an "Active Mode/Passive Mode" switch which is triggered by the computer as necessary.

Still, whatever scenarios can be imagined to question or confirm the technology, the human element is more important. I am fascinated by the technology and marvel at it, but I think the big questions here will soon not be "does it work" but "will the insurers, governments and most of all the public - in the plane and on the ground- accept it and how will they interface with it?"

[Mk 1]

TD Racer - Spot on. Do you remember the arguments that broke out when the first cars came out equipped with ABS? "How can a stupid computer know how to brake better than my (insert years) or experience?" Now people have accepted that under 99% of conditions an average driver will be better off stamping on the brake pedal and letting the ABS computer do its thing.

One of the latest Ferrari's is faster around a race circuit when the computer systems can intervene to assist the driver than with the driver doing it without aids. To prove the point that some exceptional humans can beat the microchips, apparently Michael Schumacher was the only driver to better his times with all the driver aids turned off.

There was a debate that perhaps Sully could have made it to whatever the alternate airport was, but in his judgement he could not have. That doesn't in any way damn his airmanship - he did the best with what he had (Mk 1 grey matter and years of experience)he saved a plane load of people, but a computer could have probably made a definitive calculation within seconds of the impacts rather than 15-20 or 30 seconds that Sully took. Maybe that airport was within range given the known parameters - a computer if programmed correctly would certainly have known.

Autonomous driverless cars are now a reality and being tested. Many would argue that despite being a simpler 2D (no altitude) environment that the roads are a far more chaotic environment.

I don't agree that it cannot be done, nor do I agree that people will never trust a computer. When the general public is satisfied that computers stuff up less often than pilots, then the sheeple will embrace the tech.

I personally would still like to see a pilot in an emergency centre somewhere that can evaluate the computers decisions when things do not go according to plan, and make corrections where necessary.

[Huck]

A PPrune thread on planes with no pilots aboard.

Ah yes, the semi-yearly mating call of the engineers.

Tell you what, mates. You get that 787 going for a couple of years without an inflight emergency and we'll move on to the next challenge.....

Meanwhile, I'll go fly 777 Freighters. We carry four (4) pilots. The horror!

[egravitics]

Bean-counters fly aeroplanes, in the sense that they buy them, buy the crew for them, buy maintenance for them, buy advertising for their brand of service and then put them into the skies to earn a return on the capital used. In the bean-counter's mind margin is everything, hammer down costs in the flight factory whilst finding ways to promote a premium service on board to push up fares for as many seats as possible. Concorde was profitable once the right fare was established in the consumer's mind. Concorde wasn't killed off by flight 4590's PR hit as the piece of metal on the runway wasn't Concorde's fault. Concorde died because of the underlying fundamentals of rising costs in its flight factory; flight 4590 just made it convenient to call time. Super-sonic is a margin dreamland for the premium fare and the SSJ executive jet will soon be here with sonic boom. Ramp up margins again once sonic and boom are no longer closely associated words for fast travel (Aircraft Systems Research Group : Institute of Aeronautical Technology - silent supersonic (http://www.aero.jaxa.jp/eng/research/aircraftsys/supersonic.html)).

Similarly automation and the right story for its PR will get it introduced as soon as the margin benefits appear. Will automation get fatigued, need rest/layover hotel rooms, demand changes in working practices, want pay-rises, have to belong to a Union, need multiple shifts on long legs? No. Will there be "teething problems"? Yes, in the same way the new tech of the 787 is having teething problems or the all-aluminium pressurised airframe had teething problems (De Havilland Comet) (http://en.wikipedia.org/wiki/De_Havilland_Comet). 60 years ago to the year, Comets were falling out of the sky in bits because an aspect of aluminium technology was not understood. Should aviation had stuck to the airframe technology of the Dehavilland Mosquito or worked through the problems of the Comet? By analogy there will be a point in the future when customers will only want to fly 787 type technology because the 380 type will look fuddy-duddy/quaint/bizarre/dangerous. Extrapolating there will be a time when having a human do the flying will be viewed in an equally suspicious fashion. In between there will be losses and discoveries of what collectively is not known. You can also see a bean-counter insisting that all human pilot-monitoring (of automation) roles are dual-functioned - when not needed to do "pilot-stuff" they should be on-board with another capability adding to the bottom-line, perhaps as a flight marshal or part of the hosting staff?

[Tourist]

Huck

I'm a pilot, not an engineer.

I don't argue that it will happen because I want it to, I argue that it will happen because those bloody engineer keep making planes that need less and less pilot involvement.

Many argue on here about who is responsible for the relentless fall in pilot terms and conditions, and the simple answer is that engineers are.

Every time they make the job just that little bit easier and safer, a few more people fall into the demographic of people who can do the pilot job within the legally required safety margins.

People on here who argue that pilotless aircraft will not happen because nobody will insure them/certify them have missed the point. As these systems improve, it may be that insurers will only insure pilotless planes.

parabellum

Full size or tiny toy, the technology is the same, but easier to fit to a large airframe.

MG23

I don't think you understand how software patches work.
You upload to all the aircraft in the fleet.
Job done.
You do not then have to test in each aircraft.

To have the same effect in a human pilot requires individual training and vast expense.

[Jet Jockey A4]

Never! At least for a long while.

[911slf]

OK I am not a pilot. John Farley was chief test pilot on the Harrier and knows a thing or two about aviation and automation. I'll quote something he wrote in Flight Magazine about the Hudson ditching.

He said that the failure of both engines in a place from which a runway could not be reached is beyond any plausible capability of an automatic pilot to decide what to do next. But a human, keeping his cool, could decide to instruct the system to land on the river, and once instructed, the system could in principle carry out this difficult task better than most humans would do it.

[Tourist]

Jet Jockey A4

The question is not about what you hope will happen, it's about what will happen.....


The sad thing about anonymous websites is that when autonomous aircraft do happen, and believe me they will, I will not only be unemployed, but I won't even have the chance to gloat about being right!


It has already happened in the military with autonomous UAVs and autonomous UCAVs under development and this is where the big spend comes from. Next I would guess the cargo sector goes single pilot leveraging the now commonplace technology from the military, then a low cost will successfully lobby for single pilot. Then the USAF will start flying their strat airlift pilotless worldwide. Then a cargo outfit will try autonomous between out of the way places and when the statistics show how much safer and cheaper it is, then the public will be persuaded without much difficulty.
Remember, Ryanair has passengers despite rightly or wrongly everyone suspecting they are more dangerous than other airlines. The wallet is a powerful.

[Tourist]

911slf

Yes, as agreed earlier, the one area that humans are superior to computers is in extrapolation. Using related or unrelated knowledge to work out how do deal with something completely new.

Sully was excellent and saved the day.
However, the technology in the autonomous helicopter in my previous post is capable of working out where to land a helicopter. This is no different than working out where to plant an airliner, and in terms of glide control no human can match an autopilot, and no computer would forget to press the ditching switch. (not a dig at Sully whom I have enormous respect for, just a dig at the very serious limitations of humans)


Remind me how many human pilots have screwed up and flown perfectly serviceable aircraft into the ground? Not all pilots are Sully.

It would be interesting to discover just how many accidents have been caused by the pilots ignoring instruments/autopilots/TCAS/EGPWS versus saved by the pilots ignoring same over the last 50 years.

I honestly don't know which side of the line we would currently be on, but the graph is only going to be going one way as tech improves.

[VH-UFO]

HDRW wrote; This may work in the USA but over here we're crowded enough that you'd find it hard to prang a 172 without hitting someone on the ground (OK, an exaggeration!), and there's no space to build new airfields anywhere useful.

What would be the point of taking the cargo to somewhere remote, when it's wanted at its (presumably usually urban) destination? The extra surface transport costs may well scupper the economics.

...but I don't think *anyone* has AutoTaxy, even in testing?

The reason I say remote airfields is because i work in the mining industry, and indirectly with autonomous mining trucks. These trucks do not have a problem driving around the pit, havent hit anything yet. In fact, on a site visit the Supervisor took me to a ramp, we parked on the ramp and one of these trucks came around the corner making a beeline straight at us. I have to tell you one hand was on the door handle. Yet this truck detected us, and drove straight around. Talking to the Supervisor and what i found interesting was that he trusts these trucks a lot more than one with a human in it.

Now when i say remote airfield, relating it to mining, these autonomous trucks have there own pit, they are seperated from those with humans in them. However there are auxiliary pieces of machinery (graders, dozers etc) with humans in them interacting with these trucks, and there has never been a collision caused by an autonomous truck. In fact its been the other way around, humans are the ones colliding with autonomous trucks!
So maybe they could split the airfields, one side autonomous aircraft, the other with humans, dont know.

So i dont believe ground interaction, or autonomous taxying would be a problem. However when they get into the air, thats a different kettle of fish that i will leave to you more experienced gents and gals, for i am a mere private pilot.

[DozyWannabe]

All I can say is that as a techie with a reasonable understanding of the technical hurdles involved - those alone make me believe it's unlikely to happen in my lifetime. And that's before we get on to the legal, regulatory and safety aspects!

[mross]

No manufacturer will just suddenly launch a pilotless plane for the reasons many have stated. As with all technology the change will be gradual. First will be planes with one pilot, then a pilot who only monitors and then short haul with no pilot. This is the way.

[Basil]

WHAT? A horseless carriage? Never happen!
Some may think this thread analogous to the foregoing. It ain't
Not saying that it will NEVER happen but it is further in the future than you think.

[parabellum]

Full size or tiny toy, the technology is the same, but easier to fit to a large airframe.

OK, go ahead and find a manufacturer willing to build it for you!

I see everyone is neatly skirting around the security/terrorist issue but until world terrorism that could interfere is totally secured fully automated flight won't happen.

People on here who argue that pilotless aircraft will not happen because nobody will insure them/certify them have missed the point. As these systems improve, it may be that insurers will only insure pilotless planes.

I will take a large bet with you that the insurance industry will be the last to come on board and that initial rates will be sky high, if not prohibitive. You will need to convince underwriters that the risk they take now, of an airframe falling into the centre of a big city, is greater than that of computer only flown aircraft. Good luck with that! (spent three years in the London aviation insurance industry way, way back but the principles don't change).

[CargoOne]

If you think that a software engineering team can program a computer to do what Sullenberger did then you are deluded.

Actually just to repeat what Sully did is not a problem at all for programming. Accurate 3D landscape stored in onboard memory, several pre-loaded scenarios to look for and application of aircraft performance (incl degraded etc) to that scenarios will give you much more safe and accurate plan to be executed. This is not a fantasy, this is 10-15 years old processing capabilities which do not require a cryogen cooled 100 tons installation.

It is a matter of how many other potential "out of the box" scenarios are yet unknown and therefore cannot be programmed at the current stage of AI development. Just recently discussed this subject with the people who are involved in some prototyping :) Watch this space...

[Nickbat]

I find this discussion a tad amazing.

How many of us have had a PC, laptop or mobile completely freeze: unresponsive to all inputs and only able to be brought back to life after a physical reboot? How many of us have experienced severe glitches after software upgrades?

Would YOU really fly in an aircraft where the possibility exists that the systems could freeze and you and your fellow passengers then have absolutely no control over your assured death — possibly with a few hours to prepare for the inevitable?

It's a bit like your local swimming pool dispensing with lifeguards as no-one has ever drowned. Yes, there would be great wage savings to be made, but would the public support such a measure? I think not.

Also, the analogy to automated trains is surely mistaken. Being land-based, they can only crash into other land-based objects — situations which can be managed by computers (though not 100%). What they don't have to face is gravity, which all aircraft routinely defy, but which nevertheless is constantly capable of the last laugh...

In my view, civilian pilots will ALWAYS be a necessity, even if it merely a matter of being present. :ok:

[framer]

Accurate 3D landscape stored in onboard memory, several pre-loaded scenarios to look for and application of aircraft performance (incl degraded etc) to that scenarios will give you much more safe and accurate plan to be executed
Several preloaded scenarios?
Can you expand on that a bit? In my mind it would be several million pre-loaded scenarios but I'm well aware that I am no expert in software programming so may have misunderstood.
This is a very interesting conversation. Personally I think it won't happen in the next 100 years but I'm enjoying the arguments put forward.
I would appreciate a response to my comments regarding human error:
One assumption that non- pilots make a lot in this thread is that pilots cause more problems than they create. The opposite is true. From planning stage to disembarkation we are recognising and mitigating errors made by software designers ( incorrect coding of waypoints etc) engineers, ATC, aircraft designers, meteorologists, baggage handlers, load control staff etc. Notice any similarities between these parties? They all have human input to the system, sometimes years in advance.How do we remove that human input? We can't, we would just be moving it away from the aircraft.
Pilots, like every other human make mistakes, but don't forget that we also recognise and fix them.
I am unsure how we could get away from the human error ( which is the aim) because the Human input is still there. One big difference is that the programmer or Engineer or Baggage handler isn't motivated to get things dead right by the fact that they are doing 400mph at night in poor weather like a pilot is.

[joy ride]

I remember an interesting case a few years ago: a pier on a big lake in Germany was going to be extended into a bridge right across the lake and it was programmed into the sat nav. The project was then delayed, but a driver was told by his sat nav to cross the "bridge".

Bystanders were amazed to see a new MB drive resolutely along the pier then right off the end and into the water. Luckily the driver survived and excused himself by saying that the sat nav had told him to go there.

Now almost everything these days turns out to be an Urban Myth, and perhaps this is one, but it was widely reported in the media.

However, it is an example of human fallibility:

1) the programmers had mapped something which did not exist...tech fail.

2) the driver - trusting the technology to the point of diminishing his own responsibility - drove into the water, possibly comparable to the recent Asiana incident...."pilot" fail.

Our solitary airliner pilot of the near future could just be a pre-flight checker, a voice for the passengers to hear, and an emergency back up. A whole career perhaps with little to do and little point in being there. Not a very good incentive to hone skills, and no way to stay alert or even awake for the one critical time that he or she might be needed.

[AviatorDave]

Pilots don't want pilotless aircraft for sure, but I would not put my bet on the passengers.
Time has not come yet, that is true, but technology will move on, and once it reaches a point where airliner operations without pilots will be feasible (e. g. with remote control from the ground as a backup, where one guy suffices to control multiple aircraft in case it would be required), the airlines will drool over the possible cost cuts, and a giant marketing machinery will be started, promising low fares at an unprecedented level.

And passengers will fall for it.

[Util BUS]

To quote Donald Rumsfeld, “there are some known knowns, some known unknowns, and some unknown unknowns.”

We live in a generation where bean counters and sales staff are in the driving seat and they leave the engineers to play catch-up. This gung ho style leads to much greater risks being taken with great pressure to deliver technology that is not yet mature.

The development of pilot-less airliners would most likely place in steps:

1) Current generation aircraft would need to be able to complete standard airline sectors with very little input from pilots. ACARS route uploads, CPDLC Comms, and advanced VNAV and LNAV would be required. Unfortunately thanks to a policy of just stretching and rehashing old aircraft designs that would still be a long way off especially for the 737 family with its antiquated master caution system and manual switching philosophy. Recent accidents show that automation is not yet up to scratch, once again pushing for pilots to be better trained rather than being able to relax and take it easy. If aircraft achieved this high level of automation, accidents and incidents should drop to almost nothing. This idiot proof concept would pave the way for single pilot airliners. As a great deal of investment is required I feel that this phase alone is still 20-30 years off as airliners that have yet to fly still have antiquated systems being built into them. Keep in mind however that it has been 25 years since the first delivery of the A320, with its philosophy of designing out the pilot, yet we still have crashes such as the GF320 and AF330.

2) The next stage would involve removal of one pilot with highly advanced artificial intelligence systems on board to reduce the pilots role more to that of monitoring and interfering. As this stage has the greatest amount of unknown unknowns we will probably hear about their introduction to the civil world at the earliest in 10 years time. However design issues and cost overruns will probably mean that we do not actually have a viable commercial airliner available to achieve this for another 40-50 years. It will probably take at least another 10-20 years of operation for flaws are ironed out and where intervention becomes minimal. Due to the huge costs, the most likely scenario would involve re-applied military technology, however because of the very different nature of military and civil flying almost a complete redesign will be required. The key hurdle during this stage would be technology and software, the closer they to achieving the goal the more difficult it becomes to move forward. Incidents such as the QF330 pitch flaw over Darwin would present themselves and cause major setbacks.

3) Combined with the data from the previous generation of aircraft along with redundancy and improved highly encrypted communication, at the next stage we would be likely to see ground based operators tracking 10-20 aircraft simultaneously with interference only on rare occasions. This could eventually be reduced to emergency crews based at the manufacturers headquarters. This step in my opinion would be one of the most difficult to achieve and could take another 50 years. By this point most technical issues will have been ironed out.

My view of how soon all this can happen is probably too optimistic as the biggest hurdle to achieving all this will be litigation. Therefore it could easily at least twice my estimations. In the relatively short twenty years I have been flying I find procedures and briefings are getting longer while technical data I get from the manufacturer seems to be getting more vague and irrelevant. Since I only really have a background on Boeing aircraft I can only speak about them. Everything is now more about lawyers then test pilots or engineers. But they still make a good business selling you AFM pages for weight changes or tailwind limitations etc.

Now take for example the case of actually really wanting to improve safety using technology. The technology to get accurate aircraft weights has been around for a least the last 70 years. Combine that with the computing power available for the last 30 years, then it should be feasible to mount weighing sensors on the oleo’s and have the aircraft weigh itself and calculate takeoff performance by itself with no input from the flight crew. This could have prevented the MK Airlines crash in Canada, EK incident in Australia, and SQ incident in New Zealand. So why hasn’t it been done yet? Well of course it would move liability from the airline to the manufacturer.

How many times have you heard a manufacturer admit they have a design flaw and need to recall or completely redesign a system. I am not talking about AD’s and SB’s for small changes, I am talking about Li-Ion type groundings. Hardly ever, because most of the time it is easier to blame things on a tool/rag left in the wrong place while the system quietly gets redesigned in the background.

As complexity grows it actually becomes even more necessary to have a fall guy to blame for things not working. Business can then continue as usual while ICAO introduces even more rules and hoops to jump through. Next will no doubt be a six monthly visual approach certification.

The other question has to be what incentive is there for the manufacturers. Boeing for example will probably want at least a 15 year production run on the B737max. If first delivery is in 2020 and last delivery in 2035 with a useful life of about 30 years then you will effectively be seeing a rehashed 1960’s design still flying in 2065. As for Airbus, I think with various tweaks and stretches we can probably see the A380 flying for another 70 years. After burning their fingers on delays with the A380 and B787 I think both Airbus and Boeing are unlikely to push for pilot-less airliners. They would rather have low risk of amending current designs with minimal changes. The only way I see them taking on such a project would be with a massive launch order financed upfront with clauses specifying there would be no penalties for delays. Or perhaps a new player would want to enter the market based on such a design, something that is likely to end in bankruptcy.

So I am curious why we are hearing so much about this now. It sounds like some accountants wet dream. Since we are all viewed as prima donnas who demand a lot and do very little it probably seems like a very good idea to someone sitting off in their ivory tower, disconnected from the reality of day to day operations. It actually reminds me of an airline I used to work for where the new CEO was trying to cut costs. The CFO told him the best way to do this would be to get rid of the top thirty percent of most expensive pilots, which he promptly did. This included firing the joint Flight Ops and Training post holder two weeks before the AOC was due for renewal. Needless to say the airline was grounded for six months while things were put back in order. The CEO was then fired only to have the CFO take his place.

But back to the topic of pilot-less airliners, what would be the one thing that really causes demand for aircraft like these? In my opinion only a massive worldwide pilot shortage, but even then occams razor would suggest the real solution would just be new, innovative, cheaper, ways of training pilots and it will be business as usual.

[Tourist]

"Would YOU really fly in an aircraft where the possibility exists that the systems could freeze and you and your fellow passengers then have absolutely no control over your assured death — possibly with a few hours to prepare for the inevitable?"

Millions do every year......

They are called airbus.

Any fly by wire system has theoretically got that possibility. If you don't understand that, then you are just waffling and have no business on this site.

In any such system, the pilot tells the computer what he would like the computer to do. If the computer crashes, no input will reach the aerodynamic surfaces.

There is no manual backup. If the computers are dead, there are no controls. Period.

[Tourist]

Util Bus

You are being silly.

You are picking timeframes out of your @rse with no justification whatsoever.

Look at a 50 yr old aircraft and see how much similarity it has to a 787.
Add to this the pace of technological advance and tell me that it will take another 50 to be capable of unmanned flight.


"Keep in mind however that it has been 25 years since the first delivery of the A320, with its philosophy of designing out the pilot, yet we still have crashes such as the GF320 and AF330.


Randomly irrelevant statement. Yes the 320 is 25 years old.

1. It has not been enhanced in 25 years, so you can hardly expect it to exhibit new capabilities.

2. An automated 330 would have believed the stall warning and lowered the nose thus saving the passengers. Using examples of pilot incompetence to naysay automated aircraft is a novel approach, but I admire your adventurous approach.

[Util BUS]

Tourist

Firstly, if an A320 loses both computer systems then it ends up in direct law and is still completely flyable.

Second, I stand by my time estimates based on what I wrote. Aviation hasn't changed that much since the 1980's. There seems to be a lot of tweaking going on, but no major leaps and bounds. i.e. If it ain't broke then don't fix it. This leads to concepts such as the relaunch of the twin otter program and a market where Boeing can still sell the B737.

You seem to be taking this all very personally as if you had written the article yourself. Could pilot-less airliners be available in the near future? Yes.... If we had a government funded program similar to that of the space program in the 1960's. Is this likely to happen? No... Therefore things will continue as they are because no one wants to throw billions at a pipe dream until the technology is ready.

You say you are a pilot, but you must be operating on some different planet or maybe some other aircraft type which I have never heard of, because otherwise you would agree with me.

[framer]

2. An automated 330 would have believed the stall warning and lowered the nose thus saving the passengers.
Ahh so there will be a prioritisation between systems so it will believe one over the other. So it ignores the probes and believes the AOA? What happens when the probes are right and the AOA is wrong? What happens when the radar hasn't picked up a CB growing at 5000fpm ( or it has but has prioritised the 4D traffic deconfliction over the weather ) and the Radom and AoA and probes are all damaged?
Can you answer my question at post 114 Tourist? Also my reference to the Qantas A380 ?

[Dimitris]

There is a lot of mention on the Hudson ditching.

Didn't the automation onboard made the ditching easier? I'm not talking about the 'ditch switch' but the alpha protection.

Wouldn't the touch-down be better if the alpha-protection had a sub-mode for ditching, to maintain energy and adjust for the optimal AoA on splash-down? Alpha protection was operating as designed, had it been designed to splash-down at specific AoA and RoD it would have done that pretty good, and assisted even more an excellent pilot.

The pilot is hero because he made a decision and saved the day, but how many would have called him a hero if something had gone wrong in the ditching? Half of this forum (at least) would have been arguing about turning to TEB, which the report now says that could be reachable. A computer would have made it to TEB, a good pilot in an automation augmented plane saved the day, another pilot might have not made it.

Like others said before, an average driver wont break better than the standard ABS in a car. If you are a race driver you don't need ABS. But I'm sure that even race drivers commute with ABS equipped cars.
Of course the you need to push the brake pedal, otherwise a 'perfectly serviceable' automobile crashes.
The more 'perfectly serviceable' automobiles crash, the more sense it makes to automate automobile braking by adding cameras, lidars, RF range finders etc etc.

Sorry for the ABS drift, but have been saved by ABS many times, though I've crashed because I haven't pushed the pedal when I should have.

Delete if irrelevant.

[CargoOne]

Several preloaded scenarios?
Can you expand on that a bit? In my mind it would be several million pre-loaded scenarios but I'm well aware that I am no expert in software programming so may have misunderstood.

Lesson from Sully can be formulated as "look for large water areas if other landing sites are not available", however this is hardly any news. And without any disrespect to Sully a fully automated computer system would probably choose to land in Teterboro and his aircraft will be flying again in a matter of days after engines and slides replaced and heavy landing check performed. Simply because human brain cannot calculate exact performance, descent profile and flight path for 25 different landing sites in 1 second. Computer can.

[Mk 1]

Tourist

Firstly, if an A320 loses both computer systems then it ends up in direct law and is still completely flyable.

OK, substitue A320 for a dynamically unstable FBW design - say the F-16 or just about any of the current gen airframes. No computer and you lose control in seconds. Yet we do not hear about airframes being lost on a regular basis because of computers doing the 'blue screen of death' scenario.

Redundancy, meticulous testing and planning.

[Lookleft]

Until all the computer nerds can guarantee that a change to operating software will never have an unintended consequence then there will still be two bods up front. For those who are interested in what spurious electrons can do Investigation: 200503722 - In-flight upset; 240km NW Perth, WA; Boeing 777-200, 9M-MRG (http://www.atsb.gov.au/publications/investigation_reports/2005/aair/aair200503722.aspx)

[Util BUS]

Some stats:

Northrop Grumman - Global Hawk:
Wingspan wider than that of a B737.
Projected Cost $35 million each or the price Ryanair might pay for a B737-800.
Actual Cost $218 million each or the price of a new A380.
Cost over-run of 620%.
Block 30 version was due to replace the U-2 from 2015, but has now been shelved because it costs more to operate.


United States Military Drone Programs:
Total cost of US drone program $23 Billion.
USAF had 129 medium and high altitude drone crashes worldwide in the last 15 years with damage in excess of $500,000.
Most accident prone fleet in USAF with 9.31 crashes per 100,000 hours versus 3.03 for other fixed wing aircraft.

Doesn’t exactly bode well for pilot-less airliners.

[mross]

A pilotless aircraft does not have to be 100% safe. It just has to safer than a piloted aircraft!

[framer]

That's true mross. Who is going to pay for the flying required to establish those statistics? The $23 billion that util bus talked about certainly wouldn't be available to a private company or even an Industry collaborative so which government is going to stump up with the required amount? It would be political suicide.
Still nobody has answered my question about moving the human error down the line as opposed to removing it altogether. A human has to be involved in programming, in establishing and confirming fuel, freight, pax and baggage weights etc. in positioning the weight so tat the Tomac is correct.The humans doing this will never be as invested in the outcome as a pilot who rides with the machine.
At my airline we have a recurring problem with waypoints being coded incorrectly that nobody seems to be able to get to the bottom of, this is not a technology problem, it's a human one, how do we remove this?
What happens when the met computer models get it wrong and all the heavies inbound to Perth decide ( autonomously) that Kalgoolie is the place to go but then realise ( autonomously) that here is no room on the Tarmac for them? Do they autonomously decide that parking a 777 on the grass is better than landing with winds that exceed the aircrafts capabilities?

[John Farley]

Chaps/Chapesses

Can we have a sensible debate about the future of automation that goes beyond the “It ain’t going to happen” and “yes it is” level of comment?

In order to do this we need to think about just what the current two crew members do. I suggest they do two things:

1 Operate the aircraft (which involves making and executing countless decisions about each trip between push back and shutdown)

2 Steer the aircraft (directly control any flight parameter such as speed, height, heading and so on)

So far as 2 is concerned this has become a fairly intermittent task and since it involves much training, skill and currency to do it well, doing it intermittently is producing some well ventilated problems at present.

I know from personal experience that automatics can steer (as defined above) an aircraft much better than me when they are working. However in the past I also know they can fail – just like that.

My answer to the problems inherent in 2 is to improve the automation and make it fail safe by design and thus stop the need for the crew to steer the aircraft.

My view about 1 is that this task will not go away in any foreseeable time scale (thanks to a whole host of reasons from company needs to weather, to customs, to criminality and dealing with drunks to name but a few).

So we need two people in the cockpit but I don’t want them to have to use manual skills - just their knowledge and decision making talents to operate the aircraft reliably and efficiently.

[furball_t]

The future they want to impose is machinery completely automatic with few human being checking the system in case there is a major failure.

Strange future in my opinion

The Age of Spiritual Machines: When Computers Exceed Human Intelligence: Ray Kurzweil: 9780140282023: Amazon.com: Books

[Util BUS]

John

Focussing solely on your step 2, which I would equate to step one on my previous posting about steps needed to get to a pilot-less airliner we are still faced with several big issues. Primarily these all centre around money or for that matter a lack of it.

During a time when airlines were still nationalised money could be spent on expensive national projects such as Concorde etc. Margins are thin and these days all airlines are looking for the best deal. They in turn put the screws on the manufacturers, who in turn simply go on trying stretch and rehash old aircraft. Using these grandfather rights the airframe manufacturers can get around the huge expense having to recertify the aircraft, but at the cost of having older technology on board. Therefore in civil aviation we a probably running at least 20 years behind what a home user might have on his PC.

Apart from the latest generation aircraft such as the B787 or A380 most top of the line FMC’s on civil aircraft have a 1 megabyte storage capacity with a processor speed in the Mhz and not Ghz range. This would apply to about 95% of aircraft flying worldwide. Using such antiquated technology starts to add a lot of restrictions. So just flying around in cruise the aircraft can do a reasonable job, using LNAV and VNAV, if coupled with a reliable CPDLC system you could probably get as close to an autonomous aircraft as possible.

Problems quickly start to arise when you do something as straight forward as descend. Significant changes in wind speed, direction, or temperature throw off the FMC and require intervention. Part of this is down to the simplicity of the autopilot design. Either it will use VNAV following a pre-calculated descent path which is not adapted over time to account for change in environment, or if this is not doing a good job manual intervention is required and we can use FLCH which is governed by speed, or V/S which is governed by a descent rate.

So instead let’s look at the prospect of a climb. Due to congestion in airspace such as London once again a great deal of switching is required. Often various heading and altitude changes, requiring the selection of the most appropriate autopilot mode for that change. Sometimes you might also want to manage energy such that you can easily climb shortly when being held down by other traffic. You will of course also be restricted by company SOP’s to change to V/S and limit rate of climb or descent when approaching level offs in excess of 1000 fpm to prevent level busts and spurious TCAS warnings (something the FMC not yet programmed to do).

What about the take-off and departure. Presently take-off has to be manual, and although most departures are well depicted for VNAV, the limitations of the FMC mean that they cannot fulfil multi conditional waypoints. Therefore a restriction such as turn at 800ft but not before 1nm cannot be met. Therefore there is a choice for navigation programmers to decide on 1nm or 800 ft not both.

As for landing, non precision of Cat I ILS would not be allowed, with all aircraft consistently doing autolands. This would however restrict airport capacity as they would constantly have to operate in LVO mode as well as creating extra cross and tailwind restrictions.

To top this all off, this current generation of ultra-simple FMC’s suffers from a huge amount of flaws, meaning that sometimes route length is not properly calculated, speeds are ignored, waypoints bypassed, or the system freezes. Just take a look a the long list of bulletins in the front of every FCOM 1, most have to do with the navigation system. Furthermore since there are many different suppliers from which to buy navigation data that is all also open to data corruption and errors.

So lets take a status check on where we currently are:

TAXI OUT: Manual AUTOMATION GRADE: Fail
TAKE OFF: Manual AUTOMATION GRADE: Fail
DEPARTURE: Auto AUTOMATION GRADE: Poor to Fair (Depending on Airspace)
CLIMB: Auto AUTOMATION GRADE: Fair to Good
CRUISE: Auto AUTOMATION GRADE: Good
DESCENT: Auto AUTOMATION GRADE: Fair to Poor
LANDING: Auto AUTOMATION GRADE: Fair to Fail
TAXI IN: Manual AUTOMATION GRADE: Fail

Everything apart from that marked as good requires some sort of intervention, therefore as you can see the current state of play is not that great.

So what would be required to bring this up to speed. The most important thing for the Departure, Climb, Cruise, and Descent stages would the introduction of a newer modern FMC system. This would have to be on par with todays processing speeds and storage capacity. However what will make these systems so expensive is that they will require a great deal of new programming to make them more intuitive and accurate. They would then need to be robustly tested and certified, taking an even greater deal of time and money. Just think of all the airframe and engine combinations that would need to be recertified. To tackle the takeoff and landing phases of operation extra sensors would be required to generate a perception of the outside world. The most likely would be video or infrared camera systems. Further expensive software, testing, and certification would be required to use these. In my opinion this would be of the order of 10-100 times the costs of the FMC upgrades. It should also be kept in mind that these systems could be easily spoofed by shadows, light sources, lasers, or heat sources. As for the taxi stage, that would be a huge waste of money for relatively little gain and it would pretty much be easier to mandate that airports have a network of automated taxibots to guide aircraft around.

So once again that brings us back to the question of who will pay for all this. Especially if you have 2 people sitting up front who over time can be trained to fly even more cheaply through advances in low cost simulators.

[John Farley]

Util Bus

Thanks for that.

Sorry, but I think we are probably at cross purposes.

I see most of what you are talking about as the operation of the aeroiplane and as I said that is a complex task from push back to shut down. I see no medium term solution to those issues and perhaps no long term oine either so far as automation is concerned.

My hope is that given fail safe autos the crew will never be faced with having to take over manual control of flight parameters (ie steer the thing in my book)

Fail safe autos wil never be achieved as an update to an existing type. They would need to be designed in to some future type yet to be conceived. Just as FBW had to be all those years ago.

[Util BUS]

John

If the only purpose was to prevent someone hand flying the plane then you are 90 percent there with airbus. Boeing as has there equivalent called CWS Control Wheel Steering. The issues that would arise however if you tried to fly an approach through the MCP would be a matter of fidelity,1 deg of heading change is simply too large a turn when on approach, because of this it is cringe worthy watching someone try and accomplish just that.

[John Farley]

Util BUS

I understand what you mean when you say the Airbus can do 90% of the manoeuvres required for a flight but that is only when it is fully servicable. If there is a system degredation action is required by the crew and manual flying can be involved.

This progressive handing back of the aircraft to the crew was originally designed in quite deliberately in order that the FBW ideas could be certificated and accepted by the customers.

What I mean by failsafe (or fail operationnal if you prefer) is that the total system is designed (by whatever means necessary) to swallow a failure without performance degredation leaving the crew to decide only on whether in their view the reduction in overall redundency suggests it advisable to change the destination. So for instance in the event of loss of airspeed info in the cruise the system would maintain attitude and power setting while reconfiguring itself to use an alternative source of airspeed or AoA and so on.

I know you did not mention Sully but others have. He had to use extraordinary skill because his aircraft could not help him much. I want the future aeroplanes to be able to help the crew like this:

You are a future Captain climbing out of La Guardia when both engines fail. As the operator you decide the crisis needs a landing on the Hudson. You undo the guards protecting the Glide Landing button and press it. With your knowledge of the aircraft’s gliding performance you estimate the touchdown zone on the local area map, draw the final approach track you want with your stylus, press the Glide Landing button again and thank your lucky stars that you did not have to use skill so save your aeroplane. Just knowledge.

I'm away for a bit but good to debate things with you.

[Leftofcentre2009]

Folks once talked about this regarding trains. many said it would never happen - yet many thousands do every day on the DLR.

Indeed the Northern Line and soon the Picadilly Line will be driverless (although a man will still be seen sat there as fought by the ASLEF & RMT Unions.

Those Underground lines are in the process of being resignalled and upgraded from conventional colour light signals to "Moving Bubble" signalling.

ps I work in the rail industry :ok:

[John Farley]

Leftofcentre2009

I fully accept what you say about trains but they all operate on a continuous stream of 'go' messages when all is well. As soon as anything is out of the ordinary and a 'go' message is not received they stop.

That is a simple fail stop system.

An aeroplane has to continue to go until the crew say stop. Hence the need for a fail safe or if you like a 'fail go' system. This requires considerable levels of redundancy which are not necessary for a fail stop system (witness hotel lift fail stop systems!).

I totally disagree with those people who are talking about pilotless aircraft. I want the 'pilot' to totally control what goes on during the flight (operate it) without actually having to steer the aircraft as this needs currency, training and is particularly prone to exposing human failings, weaknesses and mistakes in general.

[framer]

I would like to hear from the pro automated airliner people how they think the software would deal with a contaminated fuel situation such as that experienced by the Cathay crew in 2010. The incident is being discussed in Rumours and News.

[fireflybob]

framer, a deathly hush!!

[John Farley]

framer and fireflybob

I don’t know if I count as a pro automation chap but I don’t see what this contaminated fuel problem has to do with software. It has to do with the crew and how they think the aircraft should be operated.

People who think automation is about no flight crew in the front of an airliner have not read what is being suggested.

I am pro the flight parameters being accurately and reliably controlled by automatics that are fail safe so that the cockpit crew do not have to use skills they cannot easily maintain in today’s industry to deal with automatic failures of today’s aeroplanes where the aircraft is handed back to the crew.

Post 130 gives more details of my views.

[framer]

People who think automation is about no flight crew in the front of an airliner have not read what is being suggested.
Hey John, I have just gone back to the original post and had a re-read of the article to make sure I haven't misunderstood what is being suggested. The first paragraph is The pilotless airliner is no longer unthinkable. It is just a matter of time before airliners have one pilot, and soon after that they will have none.
My understanding is that we are discussing how long it will be until there are no pilots in the flight deck, just automated machinery controlling the flight. Can you point out where I am wrong there as I genuinely want to know if I am.
With that in mind, my question stands, how would the pilotless aircraft manage the fuel contamination situation where one engine is at 76% and the other at idle and there is no way of knowing if either will remain at those settings.
The silence has been quite deafening.
Once we have determined how the pilotless aircraft would have dealt with that situation I would like to hear how it would deal with this situation which occurred in 2012 to an Air France aircraft:
About 10 minutes after crossing latitude S12.5 the captain, pilot flying, switched his navigation display to 160nm, the weather radar scanning around a center tilt angle of -1.5 degrees, the aircraft moved in clear skies with view of stars, no returns on the weather radar. Dar es Salaam center queried whether the aircraft could climb to FL380, the crew declined due insufficient margin to the maximum cruise level possible. The aircraft was on autopilot and autothrust, LNAV, ALT and SPEED modes active. Another 6 minutes later the cruise speed 0.81 mach suddenly starts to increase, the captain reduces the ND range to 80nm, the crew notices a flash and a cloud to the right of the aircraft, but still no return on the weather radar. When the speed increased through 0.83 mach the crew selected 0.78 mach into the speed window and extends the speed brakes for about 15 seconds, the speed reduces to 0.79 mach but increased to 0.82 mach again. Another flash is observed immediately followed by severe turbulence, the "fasten seat belt" signs automatically illuminate and the autopilot trips off, the aircraft climbs sharply despite the captain issuing nose down inputs. An attempt to re-engage the autopilot results in immediate disconnection of the autopilot, autothrust automatically disconnected just prior to the first officer disengaging the autothrust system. The captain manages to level the aircraft off at FL380, then descends the aircraft back to the cleared flight level 360. The automation was reconnected and the flight continued to destination without further incident and landed safely in Paris. However, a passenger and a flight attendant had received minor injuries in the turbulence encounter.
The tricky bit in my opinion, for the pilotless aircraft is that the weather radar struggles to pick up returns from ice crystals.

[mross]

I think you are missing the point of this thread. The post is about future possibilities.

What can a pilot do about contaminated fuel that a pilotless plane would be incapable of doing in ten or twenty years time?

Any way, even if the scenario you describe defeats the pilotless plane this is not an argument against automation. If automation prevents the type of crashes currently being discussed on this forum and that leads to an OVERALL increase in aviation safety then we all gain. Just as the old anecdote of a driver being "thrown clear" from a car accident is not a valid argument against the wearing of seat belts!

[John Farley]

mross

I could not agree more your last.

framer

Sorry I should not have said people have not read what is being suggested. I should have said you have not read my post 130! That is where I start from. I hope I don't come across as arrogant but I think some of the early 'crew-less' ideas are way off the mark.