Autoland

[ZAGORFLY]

Air safety
The recent loss of the Cypriot 737

The investigation is still on and rather than re-assuring the passengers that flying is safe its fare to findings that are fueling rumors and several multi purpose speculations.
I hope that this tragic incident will not remain “smoke in the air”.

With my comment I only wish to drag your attention to direct the industry to implement “already available and not expensive” solutions to recover an airliner with its impaired or incapacitated crew. Any modern aircraft (including the 737 can perform the autoland but need to be assisted by the crew that still has lower the flaps at speed intervals and the landing gear normally 10 NM before the runway threshold.

For a fraction of the cost of a re-newly in flight entertainment system the avionics manufactures could easily develop a software to empowered the air traffic controllers (ATC) to instruct the flight management computer FMC (who give instructions to the autopilot) to be remotely trigged to arm the “Autoland mode” and perform the speed altitude restrictions Flaps setting and gear extension accordingly.

Trough the transponder of the aircraft the FMC could send automatically an emergency code to the ATC alerting that something must be wrong in the cockpit if a cabin pressure or any faulty system alarm is not handled by any human corrective interaction.

Certainly this latest tragic incident (the second if its kind) had darkened years of safety development in automation implementation research toward the possible goal of a total safety.

The technology is there…

Since then is better to remind yourself that at cruise altitude 60 seconds are pretty fast and is therefore better to pay attention to the safety briefing instead browsing the movies list into the in flight magazine.

[Blacksheep]

This highly trained and much qualified one-time avionics specialist often teases pilots that the avionics industry first eliminated the wireless operator, then the navigator and lately the flight engineer and that the co-pilot is next and then finally the captain.

Fun is fun, and we all like to poke fun at our colleagues from time to time. In reality, there is absolutely no way that the systems that are mentioned in the forgoing post could be controlled 'blind' from the ground. They are in reality very weakly integrated. How would the ATC Controller instruct the FMC to do anything? Datalink? What datalink exactly? Arm the auto-approach? OK, so what about the preceding mode changes and how do we enter the speeds into the auto-throttle? As for the FMC lowering the flaps and extending the landing gear... Ho, ho, ho yes - and Father Xmas comes down the chimney and leaves everyone a free Playstation 2 at Xmas. A tad too much Hollywood here methinks.

We are actually looking at a complete redesign and integration of the navigation and control systems of every aircraft in service today. No doubt it could be done - engineers can design and build anything they set their minds to - but no airline could ever afford to buy the modifications nor could (or would) any passenger afford the price of a ticket for a flight in such a machine. Looking at the certification programme for Concorde one can easily deduce that the certification process would be a somewhat lengthy process. Like about twice as long as the remaining useful life of the aircraft that are to be modified?

The FMC does have a function that will wake the crew up if they don't touch an FMC button for a while. I doubt if most pilots were aware of this as they don't get much chance to find out - the 'execute' key on the FMC's MCDU is the most utilised part of a modern airliner. The nearest thing to a pilotless aircraft we have today is the cruise missile and that is a one-way flight, pre-progammed from start to finish and with no intention to accompliash a safe landing at its conclusion. [I'd hate to have to interfere in the progress of a cruise missile flight. Think of the consequences if you failed]

Maybe in a couple of decades we will be getting somewhere. We don't even have driverless high-speed trains yet and surely that's something much easier to develop? In the meantime we will need a pilot if only to keep hold of the 'dead man's handle' Or FMC MCDU Execute Key to give it its proper name.

[ZAGORFLY]

Thanks for spending time to followup my mail.

you are really much qualified and technically prepared than me.

Forgive me if ask you some technical questions.

can data link send you an entire route with all the speed/altitude restrictions?

If yes (please confirm, I am not in the industry like you are)still somebody have to fisically accept the data or "purge"it.
Correct?

if I get 2 yes from you I guess that the dispatcher or the ATC can data link route 2 , you develope for us a way that the FMC will accept and execute route 2 regardless the MCU settings. Then
at proper route 2 speed restriction and altitude you design something that will trig the flap extension and then at 10NM dme from the (route 2 activated) runway you trig the gravity extension, so you save the electric motor thans should mouve the LG lever.
Autobrakes will do the rest for you.

in your opinion, how much it will cost ?

[greek-freak]

technically this is feasible, on some aircraft you have internet at high bandwidths available, the amount of data that would need to be transferred for routing/programming the FMC is minimal compared to the entertainment bandwidths available.

And data-link between aircraft and ATC will become the standard sooner or later.

I rather see a legal problem with such kind of remote aircraft operations.

[GlueBall]

. . Impractical Reality

develop a software to empower the air traffic controllers (ATC) to instruct the flight management computer FMC (who give instructions to the autopilot) to be remotely trigged to arm the “Autoland mode” and perform the speed altitude restrictions Flaps setting and gear extension accordingly

This accident is about cabin pressurization failure and about the flight crew's failure to follow trained procedures, either or both: 1- to check their oxygen masks and oxygen supply prior to flight, and 2- to don their oxygen masks in a timely manner after pressurization failure.

In this scenario, with the pilots already dead or unconcious, stefanoperer proposes that ATC be given the capability to "take control" of the airplane and get its autopilot not only to autoland on a specific runway, but also to reconfigure all autopilots in all jets to also lower the gear and flaps and to arm autobrakes and to operate reverse thrust to effect a safe full stop landing. Needless to say, at FL340 the pax oxygen supply will be good for maybe 20 minutes before pax will expire; this would necessiate also an autopilot mode of initiating an automatic emergency descent, perhaps long before ATC is appraised of the depressurization failure. The already overworked autopilot would then also have to send a signal to ATC notifying it of such failure.

Entirely conceivable, but impractical. What's next: build jets with huge deployable parachutes in case of catastrophic flight control malfunctions...?

[Seat1APlease]

There are two questions here, can it be done and would it be wise to do it.

Modern aircraft such as the A320 are already largely flown by the computer. The pilot moves a side stick but the computer will not let them overstress the aircraft, nor stall, it nor overbank it. Many of the switches are not directly wired to the equipment controlled, but send a signal to a computer to carry out the action. This could easily be done with gear and flaps.

It could certainly be possible to program the computer to take the final decision from the pilot but would it be wise to do so?

Any system can fail, be it human, electronic or procedural, and allowance has to be made for that failure but a line has to be drawn somewhere. It is perfectly possible statistically for all four engines on a jumbo to fail, so should a fifth be fitted then a sixth etc. etc. and what should happen when your new layer of automation fails which it might eventually do?

What if your system should fail and put the aircraft in danger, will the pilot have a big red handle to switch it off or will it be like HAL in the film 2001 where the computer over-rides the pilot, and that is the ultimate philosophical question.


Either the computer takes the final decision taking the pilots wishes into account, with the pilot unable to intervene, or you have the present system whereby the systems give the pilot all the information and help possible but ultimately he can override it.

[Blacksheep]

The problem isn't simply getting data up there, its dealing with it when it gets there. Aircraft systems are for economic and safety reasons not as fully integrated as many outside the industry might imagine. Landing gear, flaps, thrust reversers, auto-speed brakes and autobrake all require either manual operation or manual arming. There are good airworthiness reasons for this.

Then there's the problem with computing power. The typical desktop computer has a 30 gigabyte hard disk, 2.0 gigahertz processor and at least 256 megabytes of RAM (many belonging to flight simmers have as much as 1 gigabyte of RAM). There isn't that much computing power in the entire aircraft. A really top line latest model FMC would have at most 2 megabytes of memory and use a 486 processor (remember them?) Older 757s and 767 FMCs have only 270 kilobyte memories and are driven by 386 chips - which is why the pilots are continually putting data in manually to check for direct tracks to diversions and so on, as part of their routine flight progress monitoring. For safety and especially for certification reasons, the electronics are kept well away from the cutting edge. Or perhaps you would rather entrust your life to Microsoft XP Professional SP2?

Can anyone seriously expect an ATC controller - who isn't qualified to fly an aircraft - to take over a crippled airliner in unknown technical condition and fly it blind to a safe touch down on the runway? The Helios accident has certainly triggered a rush of bizarre scenarios on this website. Over in Jet Blast we have a crowd of PC Simulator pilots who really believe that it is possible for people who are not trained, skilled and professional airline pilots to recover and land a crippled airliner. Its just a fantasy I'm afraid; the the truth of the matter is that if both pilots die or are totally incapacitated then everyone else on board is going to die as well and there's nothing much anybody inside or outside the aircraft is going to be able to do about it.

Until the day dawns when we have the first airliner certificated for zero crew operation that is...

[ZAGORFLY]

Thank all for your attention.
I am proud that I am attracting some very valuable opinion in this highly respected forum.

The question is not to eliminate the crew or minimize the paramount importance of the Human factor in the Cockpit. The point is to have an alerting system that can communicate not only recording (FDR). A system that alert the pilots first “Hey I’am going to send a transponder distress call if you don’t correct this emergency!”
Ecam Action (in the Airbus) for instance.

Unfortunately and tragically this incident is also paying the price of the post 911 “safety” measures. The crew was sealed inside the cockpit and nobody after the barometric deployment of the pax oxygen mask had the idea to question the crew..!!!” Hallo Captain everything is ok there? “ Before 911 the door was open or accessible..
Is too stupid to die in this way to accept this kind of chain event. The crew probably run out oxygen first many assume. In Cathay Pacific Pilots check 2 times the regular flow of oxygen in the quik donning mask . Then the flight crew fail to asses the emergency, then for 15 20 minutes the entire supply of oxygen ran out by the passengers and no emergency descend occurred.

Flying with one pack inoperative is not a imitating factor bud a contributing factor would say FAA in an old aircraft with pressurization problem history I would have plan /request a lower cruising altitude and double check my OXY supply system twice.

Now news are reporting that somebody at the control disengaged the autopilot and tried to fly the plane ..without distress call? I guess the low fuel or the double flameout did in fact disconnected the autopilot and the plane left as result the holding pattern with only one final destination...

What I hope is that in the future we will be protected by what all pilots always say in impossible to occur (double incapacitation) they even eat different food at different time…So is such event when pressure cabin emergency occur and no emergency descend is initiated the FDR or the ECAM instruct the FMC to send a distress message or via transponder or put the plane direct into dive regardless the MCP settings to an hold at 10,000 some were in the database were is safe to fly.

How long will take an F16 to intercept that plane and make a visual inspection? Less than 20 minutes ?

Then if the ECAM (I mention knowing that 737 doesn’t have one) detect no action whatsoever by the crew automatically the “system” assume that nobody is in control in a cabin pressure emergency therefore: DOWN !!



Please don’t tell me that is rocket science or holliwood mentality.

[CargoOne]

Have you ever heard about hackers breaking into NASA, CIA, CityBank, VISA etc computer systems?

If there is a method to manupulate a passenger jet from the ground then you would not need any airport security any longer. Terrorists will learn the easier way to send an aircraft down...

[PPRuNe Radar]

The technology to fly an aircraft from the ground (including setting landing configuration) is already there. And unlike cruise missiles these aircraft return to land safely.

Look at the Global Hawk for example. It has flown the Atlantic and Pacific unmanned.

Global Hawk

However, in the civilian aviation world, there would be a lot of issues to address before such capability could be introduced. Operator training (would it be ATC ? or ground based ''emergency'' pilots ??), public acceptance, safety assurance, comms links .... the list is endless.

The means are there ... the will and finance are perhaps not yet.

How long will take an F16 to intercept that plane and make a visual inspection? Less than 20 minutes ?

Assuming you are 10 minutes from a fighter base. Large parts of the world are not.

[Faire d'income]

Might I respectfully suggest that the cost of designing the aircraft, world-wide coverage of the network and training the people to operate such a system, would cost about 50 years of the total profits of the entire airline industry.

[N5528P]

Honestly, I do not believe it would be that much of a problem to make it possible to fly the aircraft from the ground in cases of emergency. Especially not in A/Cs with FADEX, FBW and so on. Of course it would cost money and might open up oter security problems as described above.

I think it is a question of:
- do we want that?
- do we want to invest the money?
- how big is the additional security risk?

But what about a fail safe button like on trains?
If the pilot does not press a butten every ten minutes (just for example) the A/C holds automatically. Perhaps even in connection with the oxygen.

If there is no steering input at all after some time of the deployment of the masks, the cockpit door unlocks automatically, alarms ATC (via Mode S)....

Would that be a way?

Regards, Bernhard

[ETOPS773]

On a Fly By Wire aircraft its a matter of designing a standard procedure / protocol and designing systems integration.

The best one could hope for is ATC activating a emergency subsystem onboard the aircraft ,which would take the aircraft to the destination transmitting a special ssr code, slowing the aircraft down, calculating fuel on board to match flap and throttle setting, getting it on the ILS and stopping aircraft on runway and shutting down engines so it can be towed away.

Doing this on a A340 or 777 would be alot easier than a 767 or 737-200 as the systems are so much more mature.

[Junkflyer]

Do you really think the airlines would spend the money to do this? Also as previously noted many older aircraft such as 747 classics have no where near the avionics to do this and a retrofit would be prohibitively costly. The Global Hawk and such have been designed from the start to fly this way and will not kill 200 peple if it has a systems failure.

[Maude Charlee]

With the exception of the events of the last month or so, the greatest percentage of fatal air accidents in recent times have happened to 3rd world operators, flying old or badly maintained equipment in regions where the modern facilities that westerners take for granted are few and far between, and where airline operations leave a great deal to be desired in terms of training and procedures.

Autoland is not going to help you in these instances. Even in the west, the great expense involved and the time taken to get it right would be of little use as the Helios accident is very much a freak occurence (cause yet to be established), and there are still many more ways to die which autoland would be of no use for at all. You also need to ensure that airports have adequate facilities to support the kit onboard the a/c (lots don't), otherwise it is just another expensive piece of electronic junk.

Despite the current crop of hi-tech pilot aids, a/c with modern kit still fly into the ground/each other because of a chain of unfortunate events and human error. Systems like GPWS/TCAS will only save your skin if it correctly identifies a hazard, generates a warning in time and the pilots react to it correctly. Sure, it prevents far more accidents than we would otherwise have without it, but technology is not infallible.

[Knold]

It was the same thing after 9/11. Umpteen hundred ideas from all over about how ATC should be able to control aircrafts as a contingency using various contraptions and "secret" codes. But only good people would be allowed to use this technology see... Tossers

Something even only similar to these ideas is bound to create a heck of a lot more problems than it would solve.



The day I find something like this in my plane is the day I retire and become severely interessted in bingo and neighbourhood watch duty.

[ZAGORFLY]

crew incapacitation

Thank you for your very valuable comment.

The modern aviation is a fascinating science more I learn more I’m riveting myself into.
All the possible hazards have been secured by redundancy system. Nowadays computers double check each other against double or triple “electronic incapacitation” How come “Human “double incapacitation can still wipe out such safety measures?

I understand that different area/nationalities have different standard thus different approach to the training and safety but I earthly sympathize with that crew and passengers that went unconscious. I figure myself in that tragic event when nobody (apparently or eventually) reacted to any descend maneuver initiation for minutes before running out Oxygen without even recognize the pending tragic ending.

Such serious emergency without any human/electronic correction reaction could be easily be detected by stand by system offering to those dormant travelers more time to react /communicate regaing control bringing the aircraft down to a safe IRS position area of safe holding pattern navigation.

Nowadays as you know we tend to fly high (often at the limit of the coffin corner) to save fuel risking 20 sec.T.U.C. (FL40) or 10 sec in case of rapid decompression.
I am confident 200% on today procedures and safety equipments that allow me to survive the rare event of decompression however this tragic incident shows me that there ar area of safety that still can be implemented.

[Empty Cruise]

OK - that was the pro's and con's of an expensive system with unknown security implications.

How 'bout the simple one - train the CC to:

1) Use the radio (4 actions required)
2) Fly an A/P emerg descent (4 actions required)
3) Fly radar vectors & descend a/c further (2 actions required for each)
4) Slow the aircraft down with gear & flaps (1 table of speeds required + 2 actions for each stage of flaps)
5) Get radar vectored onto an ILS auto-land approach (6 actions)
6) Stop the aircraft after landing & shut down the engines (3 actions)

Yes, it is not for all, and yes, there are numerous risks involved. But looking at the alternetive is not all that attractive either, is it? If we assumed that we took 50% of CC to the sim to learn this (and practice a couple of times every year) - we would only have used a fraction of the dosh involved in going the ATC-controlled route - and created no further risk (what's Osamas boys gonna do - threathen to block all ATC com with an open mike in the BDX N sector or do an autoland at your airport unless we disband western civilisation within 48 hours?)

Just worth a thought - brgds fm
Empty

[ARINC]

Once mission parameters are programmed into Global Hawk, the UAV can autonomously taxi, take off, fly, remain on station capturing imagery, return and land. Ground-based operators monitor UAV health and status, and can change navigation and sensor plans during flight as necessary.
Global Hawk currently is undergoing flight testing at the Air Force Flight Test Center at Edwards Air Force Base, Calif., with more than 1,700 hours and more than 120 successful sorties flown.


The future of air transport

No crew........and blow me... it even lands itself.

Now we need good pilots soon we'll only need good engineers

[Complex_Type]

Over in Jet Blast we have a crowd of PC Simulator pilots who really believe that it is possible for people who are not trained, skilled and professional airline pilots to recover and land a crippled airliner. Its just a fantasy I'm afraid;

I've seen numerous people with no connection to aviation land in sims at their first attempt with nothing more than an instructor talking to them. Then they do it again without the prompting. I mean full flight motion certified CAE sims.

In a real situation the talk down would be available through atc comms, pilot in the tower.