Moderators: jsumali2, richierich, ua900, PanAm_DC10, hOMSaR

 
User avatar
cirrusdragoon
Topic Author
Posts: 188
Joined: Thu Feb 22, 2018 6:42 pm

Airbus completes testing of self flying jet

Mon Jul 27, 2020 10:08 pm

https://www.businessinsider.com/airbus- ... sts-2020-7

It will be interesting to see how this develops over ten , twenty years from now . Will there be a reduction in pilots onboard? Could society be persuaded to welcome a pilotless flight deck? Time will tell.
 
frmrCapCadet
Posts: 4232
Joined: Thu May 29, 2008 8:24 pm

Re: Airbus completes testing of self flying jet

Mon Jul 27, 2020 10:41 pm

It is coming. Next in line, self forming pilot's associations - computers of the world, unite. More seriously, fifty years ago there were steel walls between Certified nursing assistants, LPNs, RNs, and doctors. I see the same thing happening in flight personnel.
Buffet: the airline business...has eaten up capital...like..no other (business)
 
LCDFlight
Posts: 559
Joined: Wed Jan 01, 2020 9:22 pm

Re: Airbus completes testing of self flying jet

Mon Jul 27, 2020 10:54 pm

Airline cabin, 2050

FA, looking out the window: "Captain Siri, look at those storm clouds. Are you sure everything is OK? Why is everything so dark"

Siri: "Nothing to worry about. The automatic flying computer has never made an error. I wouldn't worry about it, if I were you."
 
AngMoh
Posts: 1051
Joined: Fri Nov 04, 2011 5:03 am

Re: Airbus completes testing of self flying jet

Mon Jul 27, 2020 11:21 pm

cirrusdragoon wrote:
https://www.businessinsider.com/airbus-completes-autonomous-taxi-take-off-and-landing-tests-2020-7

It will be interesting to see how this develops over ten , twenty years from now . Will there be a reduction in pilots onboard? Could society be persuaded to welcome a pilotless flight deck? Time will tell.


I was at a conference last October and on a panel session an Airbus R&D director (more on the R side than D side) stated that they are working towards single pilot operation for the A32x aircraft.
727 732 733 734 735 73G 738 739/ER 742 743 744/M 752 753 762 772 77E 773 77W 788 A300 A310 A319 A320 A321 A332 A333 A343 A345 A346 A359 A35K A388 DC-9 DC-10 MD11 MD81 MD82 MD87 F70 ERJ145 E170 E175 E190 E195 ATR72 Q400 CRJ200 CRJ700 CRJ900 BAE146 RJ85
 
User avatar
pilotkev1
Posts: 68
Joined: Fri Jul 22, 2005 6:53 pm

Re: Airbus completes testing of self flying jet

Tue Jul 28, 2020 12:00 am

This will be wonderful for commuting crew and dispatchers... one more 'jumpseat' and half as much competition,
 
User avatar
24Whiskey
Posts: 71
Joined: Sat Oct 25, 2014 2:05 am

Re: Airbus completes testing of self flying jet

Tue Jul 28, 2020 12:55 am

pilotkev1 wrote:
This will be wonderful for commuting crew and dispatchers... one more 'jumpseat' and half as much competition,


Except when 117 is expanded to include commutes - as it probably should've to begin with. I'd wager fatigue would increase without having someone else to talk about smoking brisket or how to startup your own rental property scheme....
 
IADFCO
Posts: 190
Joined: Sun May 22, 2016 4:20 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 3:10 am

"For its passenger jets, though, Airbus states the tech won't replace pilots in the cockpit but will make flying safer by helping reduce workload."

That for me is the key takeaway. The big obstacle to fully autonomous flight of a commercial airliner is the handling of unexpected circumstances, such as weather, aircraft failures, and traffic conflicts. One can envision potential solutions to all three, but in any case the computational power required is way beyond what is practical today. However, developing the technology is a great idea, and pieces of it can indeed greatly improve safety in the short term (i.e., in the next few years).
 
Waterbomber2
Posts: 1290
Joined: Mon Feb 04, 2019 3:44 am

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 6:21 am

If they can make cars drive autonomously, why wouldn't they be able to make aircraft fly autonomously? Less obstacles, less improvisation required.

Pilots still have to save the day almost everyday with current tech aircraft but checklists can be run by computers too and faster.
It all is a matter of how many scenarios you can program into the computer, so what OEM need to do is to collect data from as many flights as possible, and program all the scenario's so that issues can be resolved autonomously.

An onboard pilot serves as back-up, but the airplane should also be able to be flown remotely.

Pilots hate to talk about this, but the fact is that it only takes one mistake, one bad judgement, and your children and your passengers' children may never see their mommy or daddy ever again.
This is now a rare occurrence, but if we can do even better, why not?
We are only few years away from launching to Mars with hyper-automated machines, so there'll be jobs for pilots, and they'll be more meaningful and satisfying, no need to be protective.
Let's embrace progress and focus on bigger things.
 
BA777FO
Posts: 577
Joined: Sun Apr 22, 2018 2:58 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 6:53 am

Waterbomber2 wrote:
If they can make cars drive autonomously, why wouldn't they be able to make aircraft fly autonomously? Less obstacles, less improvisation required.


How will the computer know when to carry out an unannunciated checklist? Smoke, fire and fumes checklist? How will it recognise airspeed unreliable and determine which one, if any, are reading correctly? How will it determine whether it has a fuel leak or a dodgy guage? Recognise Ice Crystal Icing? Or volcanic ash? How will it determine whether it has had an engine fail or severe damage?

I have no problem talking about it but if a computer cannot sense the above, how can it deal with them?
 
FluidFlow
Posts: 703
Joined: Wed Apr 10, 2019 6:39 am

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 7:49 am

BA777FO wrote:
Waterbomber2 wrote:
If they can make cars drive autonomously, why wouldn't they be able to make aircraft fly autonomously? Less obstacles, less improvisation required.


How will the computer know when to carry out an unannunciated checklist? Smoke, fire and fumes checklist? How will it recognise airspeed unreliable and determine which one, if any, are reading correctly? How will it determine whether it has a fuel leak or a dodgy guage? Recognise Ice Crystal Icing? Or volcanic ash? How will it determine whether it has had an engine fail or severe damage?

I have no problem talking about it but if a computer cannot sense the above, how can it deal with them?


Smoke, fire and fumes checklist? Ho do you figure out there is fire in the engine or smoke/fumes in the hold?
The aircraft tells you, it already knows it before you do. The checklist? Saved on the computer/tablet you use to execute it. It is already there

How will it recognise airspeed unreliable and determine which one, if any, are reading correctly?
How do you figure out that airspeed is not correct? You read it from the displays in front of you, the aircraft already knows it, same with the checklist, it is just not programmed to execute it.

I could go on with the other mentioned things but as you already know all the information you get are actually from the aircraft and as you solve them by inputting specific actions into the aircraft you are just a middle man that can be replaced by the computer, faster and more reliable.

There are actually scenarios where the computer has its limits but interestingly none of the above mentioned are said scenarios.
 
peterjohns
Posts: 187
Joined: Fri Jan 02, 2009 11:49 am

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 9:47 am

Nearly every day, we Air Traffic Controllers have to deal with a/c with a sick passenger, unruly passenger, diversions due to many reasons. These decisions have to be made by someone responsible on board. So there will always be a crew member on board who can make and forward the decisions.

There are many efforts going on to reduce staff and workload by automizing procedures. For instance something called CDO (continous descent operations) where a plane flies a given approach profile by itself- not only the ILS but out of cruise alt.

If you look at the drone technique, automated flying is technically possible. But for stated reasons there are crew necessary, for redundant reasons at least two. So for a long time on I believe that they will be still called pilots!
 
AAY224
Posts: 11
Joined: Mon Jul 20, 2020 5:32 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 10:49 am

Not that full automation will happen anytime soon, but it can't be fun to a pilot waiting for a furlough on October 1st and then read that manufacturers are working hard to automate your job at the same time.
 
BA777FO
Posts: 577
Joined: Sun Apr 22, 2018 2:58 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 10:53 am

FluidFlow wrote:
BA777FO wrote:
Waterbomber2 wrote:
If they can make cars drive autonomously, why wouldn't they be able to make aircraft fly autonomously? Less obstacles, less improvisation required.


How will the computer know when to carry out an unannunciated checklist? Smoke, fire and fumes checklist? How will it recognise airspeed unreliable and determine which one, if any, are reading correctly? How will it determine whether it has a fuel leak or a dodgy guage? Recognise Ice Crystal Icing? Or volcanic ash? How will it determine whether it has had an engine fail or severe damage?

I have no problem talking about it but if a computer cannot sense the above, how can it deal with them?


Smoke, fire and fumes checklist? Ho do you figure out there is fire in the engine or smoke/fumes in the hold?
The aircraft tells you, it already knows it before you do. The checklist? Saved on the computer/tablet you use to execute it. It is already there

How will it recognise airspeed unreliable and determine which one, if any, are reading correctly?
How do you figure out that airspeed is not correct? You read it from the displays in front of you, the aircraft already knows it, same with the checklist, it is just not programmed to execute it.

I could go on with the other mentioned things but as you already know all the information you get are actually from the aircraft and as you solve them by inputting specific actions into the aircraft you are just a middle man that can be replaced by the computer, faster and more reliable.

There are actually scenarios where the computer has its limits but interestingly none of the above mentioned are said scenarios.


You've never flown a commercial airliner, have you? I don't say that to be rude but your condescending tone doesn't fit your ignorance.

Airspeed unreliable? I just read the airspeed? That'd kill me. Which one do I choose when the captain's PFD is telling me we're overspeeding and the FO's PFD is telling me we're in a stall? Which one will the new autonomous aircraft choose?

On the issue of smoke, the aircract can detect it in the lavatories or crew rest areas but can't detect nor determine the source of smoke from air conditioning or IFE. It cannot detect Ice Crystal Icing, nor volcanic ash.

This is the point of unannunciated checklists that's you've missed - they are unannunciated because the aircraft cannot detect, nor diagnose the problem. It's up to the crew to diagnose the problem and select the appropriate checklist(s). I've not even touched on issues where a binary computer cannot make decisions about delays, usable fuel and suitable alternatives given weather conditions.

Please don't tell me the aircraft can do things it can't. 10,000 hours of flying 737s and 777s has given me the experience to determine otherwise!
 
Opus99
Posts: 976
Joined: Thu May 30, 2019 10:51 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 11:09 am

Honestly, I’m not very comfortable with an aircraft with only one pilot talk less of an aircraft with none
 
anshabhi
Posts: 2252
Joined: Thu Oct 20, 2016 10:40 am

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 11:12 am

2 people for a machine carrying 200+ lives is too expensive for airline industry huh?

Even today, Airlines barely get any sympathies from the general public due to charging for petty things. This will cause a complete mistrust between people and airlines.
Last edited by anshabhi on Wed Jul 29, 2020 11:14 am, edited 1 time in total.
 
TheWorm123
Posts: 255
Joined: Sun Oct 06, 2019 8:29 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 11:13 am

Waterbomber2 wrote:
If they can make cars drive autonomously, why wouldn't they be able to make aircraft fly autonomously? Less obstacles, less improvisation required.

Pilots still have to save the day almost everyday with current tech aircraft but checklists can be run by computers too and faster.
It all is a matter of how many scenarios you can program into the computer, so what OEM need to do is to collect data from as many flights as possible, and program all the scenario's so that issues can be resolved autonomously.

An onboard pilot serves as back-up, but the airplane should also be able to be flown remotely.

Pilots hate to talk about this, but the fact is that it only takes one mistake, one bad judgement, and your children and your passengers' children may never see their mommy or daddy ever again.
This is now a rare occurrence, but if we can do even better, why not?
We are only few years away from launching to Mars with hyper-automated machines, so there'll be jobs for pilots, and they'll be more meaningful and satisfying, no need to be protective.
Let's embrace progress and focus on bigger things.

If a fault happens in a much-less-complicated-than-plane car, the car or the driver can instantly pull over to the side of the road or stop dead without much issue.

A plane not so simple, what happens if say a fire occurs which isn’t detected by the computers? Crashes like flight 111 and 295 have shown a crew only has 19 minutes to make it to the ground or share their fate.
B752 B753 A332 A321 B738
 
SEU
Posts: 278
Joined: Wed Mar 13, 2019 7:21 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 11:46 am

I think single pilot ops will happen in the next decade, with remote pilots working in an ATC style building on the ground, that can remotely control an aircraft should their be an emergency.
 
smokeybandit
Posts: 1379
Joined: Mon Mar 17, 2014 3:24 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 11:53 am

Autonomous cars haven't been without their fair share of serious problems
 
FluidFlow
Posts: 703
Joined: Wed Apr 10, 2019 6:39 am

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 12:18 pm

BA777FO wrote:
FluidFlow wrote:
BA777FO wrote:

How will the computer know when to carry out an unannunciated checklist? Smoke, fire and fumes checklist? How will it recognise airspeed unreliable and determine which one, if any, are reading correctly? How will it determine whether it has a fuel leak or a dodgy guage? Recognise Ice Crystal Icing? Or volcanic ash? How will it determine whether it has had an engine fail or severe damage?

I have no problem talking about it but if a computer cannot sense the above, how can it deal with them?


Smoke, fire and fumes checklist? Ho do you figure out there is fire in the engine or smoke/fumes in the hold?
The aircraft tells you, it already knows it before you do. The checklist? Saved on the computer/tablet you use to execute it. It is already there

How will it recognise airspeed unreliable and determine which one, if any, are reading correctly?
How do you figure out that airspeed is not correct? You read it from the displays in front of you, the aircraft already knows it, same with the checklist, it is just not programmed to execute it.

I could go on with the other mentioned things but as you already know all the information you get are actually from the aircraft and as you solve them by inputting specific actions into the aircraft you are just a middle man that can be replaced by the computer, faster and more reliable.

There are actually scenarios where the computer has its limits but interestingly none of the above mentioned are said scenarios.


You've never flown a commercial airliner, have you? I don't say that to be rude but your condescending tone doesn't fit your ignorance.

Airspeed unreliable? I just read the airspeed? That'd kill me. Which one do I choose when the captain's PFD is telling me we're overspeeding and the FO's PFD is telling me we're in a stall? Which one will the new autonomous aircraft choose?

On the issue of smoke, the aircract can detect it in the lavatories or crew rest areas but can't detect nor determine the source of smoke from air conditioning or IFE. It cannot detect Ice Crystal Icing, nor volcanic ash.

This is the point of unannunciated checklists that's you've missed - they are unannunciated because the aircraft cannot detect, nor diagnose the problem. It's up to the crew to diagnose the problem and select the appropriate checklist(s). I've not even touched on issues where a binary computer cannot make decisions about delays, usable fuel and suitable alternatives given weather conditions.

Please don't tell me the aircraft can do things it can't. 10,000 hours of flying 737s and 777s has given me the experience to determine otherwise!


No I did not fly a commercial airliner but I know my fair share about automatisation of processes and AI.

If you as a pilot can choose which airspeed you take between the two without any other indication (none, no visuals no nothing, then you guess). You need other indications like orientation of the aircraft, altitude, change of altitude etc. That are all informations that the computer has. Check list? Until you performed one, the computer performed 1000 of them and verified them against every data source and input.

Can you determine the source of smoke from the cockpit if there is a shortcut in the IFE system in row 32? No you would have to stand up and leave the cockpit. On the other side if you have IR sensors in the cabin the aircraft will detect the source before you even see smoke.

Do you think they will build an aircraft that can not detect everything possible that ever led to an incident?

Can you detect volcanic ash purely by looking out the window? I know it is possible under certain special circumstances but in general? No actually you get information from other computers that are fed with said information from meteorologists and geophysicists specialised in the movement of said ashes, no volcanic eruption goes undetected today. Your aircraft will know exactly if it will be even remotely close to an area where there is the possibility of ash in the air and will react accordingly.

Usable fuel, alternate routes, etc. are all informations that are no problem to process and make adjustments accordingly. A machine will know at every second of its flight which alternative airport will be a possible diversion for every single problem that might occur.

Of course the 737 and 777 can not do this and will never be capable of such actions, this aircraft are designed archaic and are from a time computers were limited by every metric. Future aircraft will be filled with sensors and computers capable of things that were unimaginable 30 years ago.

We already have auto-pilot, auto-land and auto-start capable aircraft and there is way more to come. In 20-30 years pilots will not exist any more in new designs but there will be operators making sure everything is fine. Than 10 years later this operators will be ground based and in 2075 this operators will be computers on the ground monitoring the computers in the air. Technology will not stop to advance, you are either part of it or be left behind.

In 70 years of aviation since the second world war when everything was manual to now where aircraft land themself, what do you think will be in 70 years time?
 
N766UA
Posts: 8335
Joined: Thu Jul 29, 1999 3:50 am

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 12:56 pm

I can’t wait to work from home and sleep in my own bed at night. Bring it on.
 
kalvado
Posts: 2845
Joined: Wed Mar 01, 2006 4:29 am

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 1:06 pm

BA777FO wrote:
You've never flown a commercial airliner, have you? I don't say that to be rude but your condescending tone doesn't fit your ignorance.

Airspeed unreliable? I just read the airspeed? That'd kill me. Which one do I choose when the captain's PFD is telling me we're overspeeding and the FO's PFD is telling me we're in a stall? Which one will the new autonomous aircraft choose?


What are the tools you, as a pilot, have to resolve airspeed issue - tools that computer doesn't have?
I would say 2 things - you can look out of the window (which is probably not gonna help) and you can use advanced processing algorithms (brain, or seat of the pants - whatever you want to call it).
One thing for sure - coumputer algorithms tend to improve. So does instrumentation.
Since you brought up 737, there is a ongoing EASA push for having third airspeed source on 737MAX - either air probe or sintetic value. This is, effectively, to remove seat of the pants from the equation.
BA777FO wrote:
On the issue of smoke, the aircract can detect it in the lavatories or crew rest areas but can't detect nor determine the source of smoke from air conditioning or IFE. It cannot detect Ice Crystal Icing, nor volcanic ash.

I believe it was mentioned a few times that swept wings are often invisible from the cockpit. So does IFE smoke - FA may be able to find the source, but FO wouldn't without getting out of cockpit. Not really a reason to have extra pilot on board..

My impression, honestly speaking, that one pilot is a must for forseeable future, but unloading tasks to equipment is an ongoing process - navigator, radio operator and flignt engineer jobs were eliminated over time. The best argument against single pilot operation I heard so far is how to bring a newly minted pilot up to speed in operating a large aircraft. It means taking a lot of responsibility - psychological responsibility is there, even if automation can do a lot.
 
User avatar
par13del
Posts: 10337
Joined: Sun Dec 18, 2005 9:14 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 1:30 pm

kalvado wrote:
The best argument against single pilot operation I heard so far is how to bring a newly minted pilot up to speed in operating a large aircraft. It means taking a lot of responsibility - psychological responsibility is there, even if automation can do a lot.

...and here in my opinion you revert to the old technology. In such an automated environment the individual sitting in the cockpit will be there to monitor the computers, whether it is a small or large a/c will make no difference to the computer. The additional point is that you still want the individual who is sitting in the pointy end to be responsible even after unloading all the work effort. I submit that after deploying all the required sensors all over the a/c and implementing stand by operators on the ground that the responsibility for safe operation will also be removed and placed where they have been for the last 20 years or so, in the boardrooms where the economist and financial advisors push the buttons of the humans.
 
Waterbomber2
Posts: 1290
Joined: Mon Feb 04, 2019 3:44 am

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 1:51 pm

TheWorm123 wrote:
Waterbomber2 wrote:
If they can make cars drive autonomously, why wouldn't they be able to make aircraft fly autonomously? Less obstacles, less improvisation required.

Pilots still have to save the day almost everyday with current tech aircraft but checklists can be run by computers too and faster.
It all is a matter of how many scenarios you can program into the computer, so what OEM need to do is to collect data from as many flights as possible, and program all the scenario's so that issues can be resolved autonomously.

An onboard pilot serves as back-up, but the airplane should also be able to be flown remotely.

Pilots hate to talk about this, but the fact is that it only takes one mistake, one bad judgement, and your children and your passengers' children may never see their mommy or daddy ever again.
This is now a rare occurrence, but if we can do even better, why not?
We are only few years away from launching to Mars with hyper-automated machines, so there'll be jobs for pilots, and they'll be more meaningful and satisfying, no need to be protective.
Let's embrace progress and focus on bigger things.

If a fault happens in a much-less-complicated-than-plane car, the car or the driver can instantly pull over to the side of the road or stop dead without much issue.

A plane not so simple, what happens if say a fire occurs which isn’t detected by the computers? Crashes like flight 111 and 295 have shown a crew only has 19 minutes to make it to the ground or share their fate.


I think that you chose the wrong examples because pilots did not change the outcome of the accidents you are mentioning.
In fact, in both cases, the flight crew likely became incapacitated before they could attempt even an off-field landing.

Some people think that it's all about saving on pilot salaries.
Have you ever considered how much prime real estate the flight deck occupies on an aircraft? Changing the flight deck concept to grow the passenger cabin would result in huge savings in cabin space but also in aerodynamics.
In addition, in this day and age, aircraft have become so well engineered that the leading cause of accidents have become pilot error or the pilots contributed in some capacity.

The fact is that the computers can also detect abnormalities through sensors or program logics.
They can then execute corrective action faster than it takes a pilot to pull up the QRH or start executing memory checklists.

Even situations like the Sully ditching, a computer can be programmed to calculate an alternative landing site within seconds, so that it would be able to know whether TTB is within gliding range, calculate glide angle and configuration based on obstacle clearance for bridges using detailed EGPWS data and such and read-out pre-programmed off-airport landing sites.

If a car can be programmed to stop for a child crossing the street unannounced, I don't think there is much that a pilot's eyes can do that a computer can't do.

I hope that pilots embrace progress so that BA777FO can some day change his pseudonym to SpaceCaptain and worry about launching people to Mars. In one century, we have conquered the skies, we're half-way a century where we conquer space.
Don't fight progress, be part of it.
 
IFlyVeryLittle
Posts: 112
Joined: Mon Jan 29, 2018 7:31 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 2:26 pm

Unless and until a labor union representing computer pilots becomes more powerful than a labor union representing actual pilots, you will not see autonomous airliners. Likewise, in large part, autonomous trucking.
 
User avatar
OA940
Posts: 1990
Joined: Fri May 20, 2016 6:18 am

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 2:29 pm

I think we're really ignoring the point someone made above that this won't be used for single pilot ops in the forseeable future, instead to reduce workload for the TWO pilot cockpits. We're at least 20-30 years away from single pilot ops, and way more than that (if at all) from fully autonomous aircraft. I mean, simply from a technological standpoint we're still quite a ways off. System malfunctions on planes happen all the time, so what will the computer do if it is the one that's malfunctioning? Or, how will it cope with pilot incapacitation or a medical emergency in the cabin? The issue most of us have with this idea is that the tech needed for single pilot ops is far from what we have now and the number of failsafes required makes it a very complex chain that could collapse easily. Sure in 2050 we may have progressed far enough so that it can happen but until then it can't be acheived with the same level of safety.
A350/CSeries = bae
 
olle
Posts: 2258
Joined: Tue Feb 06, 2007 3:38 am

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 2:34 pm

I suppose that freight both new built ones and P2F will first come to mind.

How complicated would it be to make a P2F A320 or A330 single pilot operated?

I can assume that this would give a huge advantage financially. And we do not talk about 200+ pessenger either at least not for a few years.
 
kalvado
Posts: 2845
Joined: Wed Mar 01, 2006 4:29 am

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 2:55 pm

olle wrote:
I suppose that freight both new built ones and P2F will first come to mind.

How complicated would it be to make a P2F A320 or A330 single pilot operated?

I can assume that this would give a huge advantage financially. And we do not talk about 200+ pessenger either at least not for a few years.

For cargo, the drawback is that there are no FAs who can assume minimal backup duties (e.g. press Big Red Button if anything happens to initiate emergency autoland)
 
Gremlinzzzz
Posts: 90
Joined: Fri Jan 24, 2020 4:28 am

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 3:07 pm

anshabhi wrote:
2 people for a machine carrying 200+ lives is too expensive for airline industry huh?

Even today, Airlines barely get any sympathies from the general public due to charging for petty things. This will cause a complete mistrust between people and airlines.
If every industry was not a race to the bottom, then airlines would not need to charge for petty things.If unions were not always pushing for higher pay, or management taking money out instead of paying down debt.........you get the drift, then maybe we would get more healthy airlines.

Anything that is repetitive is going to go the way of automation. There used to be a flight engineer, he was kicked out of the cockpit because of automation. There used to be a navigator, kicked out because autopilot became extremely reliable. You used to go to travel agents, today you can bypass them, you had to have someone check you in, and if you look at some terminals, they are going almost full automation e.g. Changi Terminal 4.

These OEM's will look to automate anywhere they can because this is what will sell to airlines. We will not one man planes because everything in aviation is built around redundancy, but there will be consistent gains made in automation and machine learning not only in the cockpit, but all the way to how aircraft are serviced.
 
User avatar
TVNWZ
Posts: 2268
Joined: Fri Feb 17, 2006 9:28 am

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 3:22 pm

The Air Force is already precision flying every day with nobody in the craft.
 
Gremlinzzzz
Posts: 90
Joined: Fri Jan 24, 2020 4:28 am

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 3:23 pm

TVNWZ wrote:
The Air Force is already precision flying every day with nobody in the craft.
They do not have to worry about what passengers think.
 
144modeller
Posts: 39
Joined: Tue Sep 27, 2016 9:52 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 3:25 pm

Just a few thoughts.

Can the on-board computer recognise a light aircraft pulling on to the runway just as your airliner is about to touch down?

When following the link in the OP, I came across another headline which said that completely autonomous cars are several decades away. I know it's only a journalist's opinion, but I'm sure it's quicker and less expensive to develop a car than an airliner.

Mechanical and electrical parts are known to suffer unexpected failures, and computers can develop glitches at a moment's notice, so it would be unwise to rely on sensors, or the software or hardware that processes the info from those sensors, to solve every problem.
Performing any task can be routine as long as everything runs smoothly. Where intelligence is required is when the unexpected happens. Computers are not AI. They can only make decisions that have been programmed into them, no matter how clever the scientists or A.netters imagine them to be, and the programming depends on what the programmer can anticipate. We know from the 737MAX accidents that the computer didn't work out how to handle the problem, and apparently prevented the pilots from saving the day.

Remember the 'millenium bug'? It was the concept that at midnight on 31st December 1999, the two-digit storage of the year would switch over from 99 to 00, causing computers to behave as if their expected tasks were still a century away, that pensioners were no longer entitled to their monetary benefits, hospital patients would stop receiving automatic doses of medicine, and a host of other problems that would take too long to relate, including causing aircraft to fall out of the sky. These problems were not anticipated when the computers were designed, but fortunately somebody thought about it in time and disaster was averted. My employers dedicated a whole department for five years or more, to ensuring every individual record in the company's database (more than ten thousand employees and millions of customers) included four-digit year ID. On its own, a computer could not cope with someone apparently going from nineteen years old to minus eighty.
Personally, I was grateful that the date 9/9/99 didn't throw up an error code nearly four months earlier. Did you think of that? Be honest.
 
BA777FO
Posts: 577
Joined: Sun Apr 22, 2018 2:58 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 3:38 pm

FluidFlow wrote:
If you as a pilot can choose which airspeed you take between the two without any other indication (none, no visuals no nothing, then you guess). You need other indications like orientation of the aircraft, altitude, change of altitude etc. That are all informations that the computer has. Check list? Until you performed one, the computer performed 1000 of them and verified them against every data source and input.


I would nevet guess which airspeed to believe in the event of an unreliable airspeed situation - all indications could be wrong! That's the point. An aircraft can still be stalled with a 0 degree pitch attitude. In a situation like this you cannot be sure your altimeter is correct either, until you get to 2500RA and you have a rad alt reading. So all of your airspeed and altimeter readings are junk, what's your autopilot going to do when it happens at 30W across the Atlantic? A computer can choose whether to accept or reject a reading/output but it cannot troubleshoot in this instance when it has no reliable inputs and determine a future course of action. That's what we're there for.

[Quote=FluidFlow]Can you determine the source of smoke from the cockpit if there is a shortcut in the IFE system in row 32? No you would have to stand up and leave the cockpit. On the other side if you have IR sensors in the cabin the aircraft will detect the source before you even see smoke.[/quote]

This goes back again to my point about your experience not operating an aircraft - no, I don't determine the source. But you would understand the process and wouldn't make ignorant comments if you knew the SFF checklist and the firefighting procedure. The cabin crew report the fire/smoke/fumes to us and if the source isn't immediately obvious then the smoke/fire/fumes checklist goes through the procedure to try and determine it and isolate it. Now this checklist could conceivably be initiated by the cabin crew and it could ask them about changes to severity but they're unlikely to be able to decide with any level of competency whether a diversion is required or whether smoke removal becomes to overriding requirement. If you can determine with any degree of accuracy where an air conditioning fire or concealed electrical fire is coming from, or where the source of a fumes event is coming from, they'd already tell us.

FluidFlow wrote:
Do you think they will build an aircraft that can not detect everything possible that ever led to an incident?


As above - it may be able to detect its airspeed readings mismatch, but it cannot with any degree of accuracy confirm which one is correct or if none of them are, and what course of action to take in that event.

[Quote=FluidFlow"]Can you detect volcanic ash purely by looking out the window? I know it is possible under certain special circumstances but in general? No actually you get information from other computers that are fed with said information from meteorologists and geophysicists specialised in the movement of said ashes, no volcanic eruption goes undetected today. Your aircraft will know exactly if it will be even remotely close to an area where there is the possibility of ash in the air and will react accordingly.[/quote]

Humans have 5 senses, a computer cannot see, smell or taste. Smell, taste and visual clues are used to determine encounters with volcanic ash in conjunction with possible cues from various engine parameters. But the engines on their own will be insufficient for a diagnosis. No, volcanic eruptions don't go undetected, but ash in parts per million that may cause engine issues may not be completely accurately mapped. So you say the aircraft can be told to react accordingly - back up to the airspeed unreliable issue now your pitot tubes and static ports are blocked with ash. Good luck - you'll need it.

Usable fuel, alternate routes, etc. are all informations that are no problem to process and make adjustments accordingly. A machine will know at every second of its flight which alternative airport will be a possible diversion for every single problem that might occur.


Weather radars are not bad, but they're not perfect. I've seen out of the window some nasty CBs that weather radars simply don't or can't detect. An autopilot can determine whether or not to fly the approach given possible CB or windshear activity? It can determine when it arrives about 100nm from Orlando, with a squall line stretching from Tampa to Jacksonville, exactly what to do? It may know it has the fuel to divert to Atlanta now, but if the TAF says ATL will have +TSRA by the time you'd get there, now what? You'll end up with a lot of fuel exhaustion issues, windshear encounters or unnecessary diversions.

Fluidflow wrote:
Of course the 737 and 777 can not do this and will never be capable of such actions, this aircraft are designed archaic and are from a time computers were limited by every metric. Future aircraft will be filled with sensors and computers capable of things that were unimaginable 30 years ago.


That may be so. And that's fine for rule based decision making such as TCAS and EGPWS but not so easy when the decision isn't binary.

You may have faith in overcoming the non-binary, unannunciated situations but I doubt it'll happen in my lifetime. They're delivering A350s and B787s with 30 year life spans that are no way near it.
 
ewt340
Posts: 1271
Joined: Tue Jul 10, 2012 7:22 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 3:44 pm

I always wonder. Would pilots agree to fly an aircraft by themself (plus computer) if their wages increase by 50%?

So, Airlines save 50% of wages by only using 1 pilot and the pilots would get 50% raise.
 
BA777FO
Posts: 577
Joined: Sun Apr 22, 2018 2:58 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 3:51 pm

kalvado wrote:

What are the tools you, as a pilot, have to resolve airspeed issue - tools that computer doesn't have?
I would say 2 things - you can look out of the window (which is probably not gonna help) and you can use advanced processing algorithms (brain, or seat of the pants - whatever you want to call it).
One thing for sure - coumputer algorithms tend to improve. So does instrumentation.


Pitch and power settings. Becoming VMC is actually a big help, as is an ILS signal. Let's say the autopilot can go against everything it would normally do and determine to ditch its airspeed readings and even just fly according to pitch and power, how will it determine how/when to divert? At 30W is it really going to divert to KEF or YYT with 40 knot crosswinds and BLSN? How will it cope with a NAV AIR SATA SYS failure? Some decisions aren't binary and algorithms can't predict or determine the correct response in a cascading failure scenario.
 
TheWorm123
Posts: 255
Joined: Sun Oct 06, 2019 8:29 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 3:56 pm

Waterbomber2 wrote:
TheWorm123 wrote:
Waterbomber2 wrote:
If they can make cars drive autonomously, why wouldn't they be able to make aircraft fly autonomously? Less obstacles, less improvisation required.

Pilots still have to save the day almost everyday with current tech aircraft but checklists can be run by computers too and faster.
It all is a matter of how many scenarios you can program into the computer, so what OEM need to do is to collect data from as many flights as possible, and program all the scenario's so that issues can be resolved autonomously.

An onboard pilot serves as back-up, but the airplane should also be able to be flown remotely.

Pilots hate to talk about this, but the fact is that it only takes one mistake, one bad judgement, and your children and your passengers' children may never see their mommy or daddy ever again.
This is now a rare occurrence, but if we can do even better, why not?
We are only few years away from launching to Mars with hyper-automated machines, so there'll be jobs for pilots, and they'll be more meaningful and satisfying, no need to be protective.
Let's embrace progress and focus on bigger things.

If a fault happens in a much-less-complicated-than-plane car, the car or the driver can instantly pull over to the side of the road or stop dead without much issue.

A plane not so simple, what happens if say a fire occurs which isn’t detected by the computers? Crashes like flight 111 and 295 have shown a crew only has 19 minutes to make it to the ground or share their fate.


I think that you chose the wrong examples because pilots did not change the outcome of the accidents you are mentioning.
In fact, in both cases, the flight crew likely became incapacitated before they could attempt even an off-field landing.

Some people think that it's all about saving on pilot salaries.
Have you ever considered how much prime real estate the flight deck occupies on an aircraft? Changing the flight deck concept to grow the passenger cabin would result in huge savings in cabin space but also in aerodynamics.
In addition, in this day and age, aircraft have become so well engineered that the leading cause of accidents have become pilot error or the pilots contributed in some capacity.

The fact is that the computers can also detect abnormalities through sensors or program logics.
They can then execute corrective action faster than it takes a pilot to pull up the QRH or start executing memory checklists.

Even situations like the Sully ditching, a computer can be programmed to calculate an alternative landing site within seconds, so that it would be able to know whether TTB is within gliding range, calculate glide angle and configuration based on obstacle clearance for bridges using detailed EGPWS data and such and read-out pre-programmed off-airport landing sites.

If a car can be programmed to stop for a child crossing the street unannounced, I don't think there is much that a pilot's eyes can do that a computer can't do.

I hope that pilots embrace progress so that BA777FO can some day change his pseudonym to SpaceCaptain and worry about launching people to Mars. In one century, we have conquered the skies, we're half-way a century where we conquer space.
Don't fight progress, be part of it.

I chose those examples because of how little time it takes for a fire to destroy an aircraft, the effect would’ve been the same of those planes had an AI pilot.

I am one of those who thinks those is to save on pilot salaries, because #1 pilots and training are expensive and #2 everything always comes down to profit and the bottom line same as 1.

The amount of extra passenger space that could be saved without a flight deck is also negligible because that’s taking into the small cockpit discounting that you’d still need space for the toilet and galley usually found directly behind the cockpit and door space. Plus you can’t sit passengers in the nose under current regulations.
B752 B753 A332 A321 B738
 
144modeller
Posts: 39
Joined: Tue Sep 27, 2016 9:52 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 4:05 pm

The post about the air force flying pilotless craft made me think. Drones are unoccupied, but they have pilots elsewhere. Having the pilot safe on the ground somewhere, but still in full control, solves a lot of problems. Fatigue or sudden illness will never be a disaster, as understudies can be on hand. No need to fly out a new crew, no overstepping the mark on working hours. One post suggested that abandoning the flight deck makes room available for more passengers. Hijackers can't hold a gun to the pilot's head. The computer doesn't have to be small and light enough to be carried.

Of course, nothing can replace the "man on the spot". And I have no info on how the air forces maintain reliable communication between drones and the ground.
 
AndoAv8R
Posts: 51
Joined: Thu Feb 06, 2020 4:29 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 4:50 pm

I think regarding full automation look at the fact how even today (with the exception of short distance transports such as the people movers in airports like Atlanta) the railroad industry worldwide still has at least one engineer at the helm both passenger and full freight locomotives that are over a mile long. And they have a lot fewer variables of things that could go wrong.
 
kalvado
Posts: 2845
Joined: Wed Mar 01, 2006 4:29 am

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 5:02 pm

BA777FO wrote:
kalvado wrote:

What are the tools you, as a pilot, have to resolve airspeed issue - tools that computer doesn't have?
I would say 2 things - you can look out of the window (which is probably not gonna help) and you can use advanced processing algorithms (brain, or seat of the pants - whatever you want to call it).
One thing for sure - coumputer algorithms tend to improve. So does instrumentation.


Pitch and power settings. Becoming VMC is actually a big help, as is an ILS signal. Let's say the autopilot can go against everything it would normally do and determine to ditch its airspeed readings and even just fly according to pitch and power, how will it determine how/when to divert? At 30W is it really going to divert to KEF or YYT with 40 knot crosswinds and BLSN? How will it cope with a NAV AIR SATA SYS failure? Some decisions aren't binary and algorithms can't predict or determine the correct response in a cascading failure scenario.

And are you sure you decision on KEF vs YYT is the best one? (nothing personal, "you" = human pilot). If you made it to the runway, it was probably a good one anyway. The difference in terms of risk is probably not that major if you can fly another 2 hours to diversion point. Yes, regulations require minimal flight time and all that, but how big it actually is? You wouldn't hesitate between 1 and 3 hours flight time, it may be 95 vs 90 minutes where you would think twice. As for the flight path taking wind into account - computer may have a wind map, and I would certainly do a fine job calculating the times.

As for the best option... Did you ever had to look back at your own actions and say "well, I would be much better off doing things differently.." ? I certainly did, both personally and professionally.

Moral of the story - there are situations where human brain is better than machine, but those become fewer as time goes by. I wouldn't look for examples of human superiority in situations covered by checklists. It is unknown unknowns when humans are better. And it turns out that these are the situations where human can screw up.
 
User avatar
SheikhDjibouti
Posts: 2228
Joined: Sat Sep 30, 2017 4:59 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 5:20 pm

BA777FO wrote:
kalvado wrote:
What are the tools you, as a pilot, have to resolve airspeed issue - tools that computer doesn't have?

Pitch and power settings. Becoming VMC is actually a big help, as is an ILS signal. Let's say the autopilot can go against everything it would normally do and determine to ditch its airspeed readings and even just fly according to pitch and power, how will it determine how/when to divert?
Exactly the same way you do, only with fewer errors.
At 30W is it really going to divert to KEF or YYT with 40 knot crosswinds and BLSN?
Way, way, way before the aircraft takes off, a programmer will have discussed the thought processes that would go through your head, and those that go through another 99 pilots heads, in order to assess the best pathway, and instruct the computer to do exactly that. Only faster.

How will it cope with a NAV AIR SATA SYS failure?
The same way you do. (sorry if it's a boring answer)

Some decisions aren't binary and algorithms can't predict or determine the correct response in a cascading failure scenario.
Yes they are, and yes they can. :banghead:
A chess game could be thought of as a cascading failure situation. With each new move, there are new possibilities for failure, and new solutions for how to survive.
And guess what game was originally used as the benchmark for computing power? :wave:

AI (Artificial Intelligence) in the cockpit is mostly about getting a computer to execute the same response as would a whole committee of the best pilots, on their best day, with lots of time to consider each problem.
Nothing to see here; move along please.
 
FluidFlow
Posts: 703
Joined: Wed Apr 10, 2019 6:39 am

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 5:33 pm

BA777FO wrote:
BA777FO wrote:
BA777FO wrote:

I would nevet guess which airspeed to believe in the event of an unreliable airspeed situation - all indications could be wrong! That's the point. An aircraft can still be stalled with a 0 degree pitch attitude. In a situation like this you cannot be sure your altimeter is correct either, until you get to 2500RA and you have a rad alt reading. So all of your airspeed and altimeter readings are junk, what's your autopilot going to do when it happens at 30W across the Atlantic? A computer can choose whether to accept or reject a reading/output but it cannot troubleshoot in this instance when it has no reliable inputs and determine a future course of action. That's what we're there for.

[Quote=FluidFlow]Can you determine the source of smoke from the cockpit if there is a shortcut in the IFE system in row 32? No you would have to stand up and leave the cockpit. On the other side if you have IR sensors in the cabin the aircraft will detect the source before you even see smoke.


This goes back again to my point about your experience not operating an aircraft - no, I don't determine the source. But you would understand the process and wouldn't make ignorant comments if you knew the SFF checklist and the firefighting procedure. The cabin crew report the fire/smoke/fumes to us and if the source isn't immediately obvious then the smoke/fire/fumes checklist goes through the procedure to try and determine it and isolate it. Now this checklist could conceivably be initiated by the cabin crew and it could ask them about changes to severity but they're unlikely to be able to decide with any level of competency whether a diversion is required or whether smoke removal becomes to overriding requirement. If you can determine with any degree of accuracy where an air conditioning fire or concealed electrical fire is coming from, or where the source of a fumes event is coming from, they'd already tell us.

FluidFlow wrote:
Do you think they will build an aircraft that can not detect everything possible that ever led to an incident?


As above - it may be able to detect its airspeed readings mismatch, but it cannot with any degree of accuracy confirm which one is correct or if none of them are, and what course of action to take in that event.

[Quote=FluidFlow"]Can you detect volcanic ash purely by looking out the window? I know it is possible under certain special circumstances but in general? No actually you get information from other computers that are fed with said information from meteorologists and geophysicists specialised in the movement of said ashes, no volcanic eruption goes undetected today. Your aircraft will know exactly if it will be even remotely close to an area where there is the possibility of ash in the air and will react accordingly.


Humans have 5 senses, a computer cannot see, smell or taste. Smell, taste and visual clues are used to determine encounters with volcanic ash in conjunction with possible cues from various engine parameters. But the engines on their own will be insufficient for a diagnosis. No, volcanic eruptions don't go undetected, but ash in parts per million that may cause engine issues may not be completely accurately mapped. So you say the aircraft can be told to react accordingly - back up to the airspeed unreliable issue now your pitot tubes and static ports are blocked with ash. Good luck - you'll need it.

Usable fuel, alternate routes, etc. are all informations that are no problem to process and make adjustments accordingly. A machine will know at every second of its flight which alternative airport will be a possible diversion for every single problem that might occur.


Weather radars are not bad, but they're not perfect. I've seen out of the window some nasty CBs that weather radars simply don't or can't detect. An autopilot can determine whether or not to fly the approach given possible CB or windshear activity? It can determine when it arrives about 100nm from Orlando, with a squall line stretching from Tampa to Jacksonville, exactly what to do? It may know it has the fuel to divert to Atlanta now, but if the TAF says ATL will have +TSRA by the time you'd get there, now what? You'll end up with a lot of fuel exhaustion issues, windshear encounters or unnecessary diversions.

Fluidflow wrote:
Of course the 737 and 777 can not do this and will never be capable of such actions, this aircraft are designed archaic and are from a time computers were limited by every metric. Future aircraft will be filled with sensors and computers capable of things that were unimaginable 30 years ago.


That may be so. And that's fine for rule based decision making such as TCAS and EGPWS but not so easy when the decision isn't binary.

You may have faith in overcoming the non-binary, unannunciated situations but I doubt it'll happen in my lifetime. They're delivering A350s and B787s with 30 year life spans that are no way near it.


Look I respect what you are doing but all you do is defend your job on a 777, an archaic aircraft that needs a pilot to function.

Pilotless aircraft will not be built that way. Airspeed can be calculated in a lot of different ways. Right now an aircraft uses one possible option. You can add multiple additional options to determine exact airspeeds independent of pitot tubes. GPS for example or active radar, passive radar from ground stations, etc. If you pitot tube fails, so what other sources still work. Same for the altimeter and all the other variables. The redundancy of sensors on „modern“ airliners is actually really bad.

You even say it yourself, the crew will help you determin where and if there is a fire. Modern smoke detectors can tell you exactly what kind of smoke there is and with IR sensors you find the exact location of the fire immediately and the correct actions can be taken. That technology already exists.

Also you make binary decisions aswell even if you do not think so, you either do something or not. Even increasing airspeed by just one knot is a binary decision.

Also for the thunderstorm szenario. That is actually one of the problems as the nonlinearity and unpredictability of storms are hard to predict but future aircraft will communicate with each other so the aircraft ahaed will warn yours and so on. On top of tjat ground stations will also transmit live data constantly. Future aircraft will have million times more data available than current ones because the computer have the strenght to actually process that information.

On the other side your ashes problem is a non-starter as any area that even remotely has the possibility of increased ash concentration will be avoided. Remember Eyafjawhtever? Within days there were non-fly zones based on computer models and measurements so no pilot would fly into it. Why would a computer do this?
 
User avatar
Airbuslightie
Posts: 4
Joined: Wed Jul 29, 2020 1:13 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 5:39 pm

Proof of concept is a great but I suspect this project is going to fall short on real world integration. Here are but a few examples :-

1.A380 cannot fly an RNAV approach after flying through an area of GPS jamming until a ground reset of the MMR's has been completed. This means that real world avionics software is falling well short of what is required. Same aircraft currently has a bug with the localiser intercept.......a feat of technology that must be 50 years old, and they still can't get it right.
2. An autonomous aircraft is going to require special ATC systems to integrate with. That is going to come at a cost and I am unsure who is going to bear that cost. The introduction of autonomous aircraft is unlikely to increase the flow rate at an airport, so I can't see massive motivation for airports to invest in this type of infrastructure.
3. To further illustrate point 2, I have been flying GLS approaches(with autoland) reliably for over 10 years now, yet this technology that can clearly benefit any airport(especially in low viz conditions) has not expanded beyond about half a dozen airports. That implies that even if all the real work issues can be sorted out, the roll out is going to occur at a snail's pace(think decades).
4. There are still parts of the planet where communications are difficult. Scope of operation would therefore have to be limited. I'm not sure how much appeal that will carry with operators.
5. It has been mentioned, but there are still several checklists that have to be activated manually due to the aircraft's inability to detect the relevant fault. Say what you like about AI but until an aircraft can intelligently & accurately initiate an emergency evacuation, there needs to be a responsible person on board.

A clinical trial flight is great but lets see this thing operate into Lagos and then decide if it has a future. I am simply not convinced that the benefits will outweigh the costs.
 
User avatar
SheikhDjibouti
Posts: 2228
Joined: Sat Sep 30, 2017 4:59 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 6:07 pm

144modeller wrote:
Can the on-board computer recognise a light aircraft pulling on to the runway just as your airliner is about to touch down?
Why not? The computer has no eyes of it's own; it's just a box full of electronics. But it will be connected to as many pairs of "eyes" as deemed necessary. One pair of sensors for clear daylight, another pair optimized for night-vision, some infra-red, some based on radar, others on Lidar. And just maybe a pair of human eyes in the Control Tower, with a hand on the big red abort button.

Mechanical and electrical parts are known to suffer unexpected failures, and computers can develop glitches at a moment's notice, so it would be unwise to rely on sensors, or the software or hardware that processes the info from those sensors, to solve every problem.
Unwise to rely on sensors. :rotfl:
And yet that is exactly where we are at today. In fact pilots have been relying on sensors more or less since 1903. Even the Wright Brothers had a piece of cloth to alert them to sideslip.

Where intelligence is required is when the unexpected happens. Computers are not AI. They can only make decisions that have been programmed into them, no matter how clever the scientists or A.netters imagine them to be, and the programming depends on what the programmer can anticipate.
No.
The programmer doesn't even need to be a pilot (although it might help), and he doesn't need to anticipate anything.
What he does need to do is take advice from real world pilots, and examine the database for crashes, and near crashes. And a thousand other possibilities. In fact "he" needs to be a multi-disciplinary team.

Remember the 'millenium bug'?
Yep, it was huge con; I know - I lectured on it.

My employers dedicated a whole department for five years or more, to ensuring every individual record in the company's database (more than ten thousand employees and millions of customers) included four-digit year ID. On its own, a computer could not cope with someone apparently going from nineteen years old to minus eighty.
Your company was taken for a ride.
If your computers / software were so old that they fell over at midnight 1999, they were long overdue for replacement.
By 1995, any company with more than two employees and a kettle for making tea already had equipment that was Millenium Bug proof.
Nothing to see here; move along please.
 
User avatar
Airbuslightie
Posts: 4
Joined: Wed Jul 29, 2020 1:13 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 6:14 pm

FluidFlow wrote:

That technology already exists.


The existence of a technology means nothing until you can find a financially viable way to incorporate it.

As an example, in the plane I fly, I cannot see the wingtips. The technology exists for parking sensors that would show me any threats to my way forward. Do you think I have this stuff on my plane? No, because somebody has to pay for it and my airline does not want to foot that bill. This is a world class ME based airline that is not short of money(pre-covid of course).

You also talk of data networks transmitting weather data to planes which is nice, and probably doable, but once again, who is going to pay for it. I am sure most problems will find a solution but you must surely see how the incremental cost for each step is making it very difficult for this to be financially viable overall. Data transmission is expensive and they will have to find multiple redundant ways to transmit the data. Then there is the cost of protecting the data stream from hackers.....maybe Garmin can offer a solution.
 
lhrnue
Posts: 366
Joined: Sat Jun 05, 2010 2:47 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 7:09 pm

smokeybandit wrote:
Autonomous cars haven't been without their fair share of serious problems


I would say the control system of an autonomous aircraft needs make less and less complex decision compared to a car.
 
BA777FO
Posts: 577
Joined: Sun Apr 22, 2018 2:58 pm

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 7:59 pm

kalvado wrote:
BA777FO wrote:
kalvado wrote:

What are the tools you, as a pilot, have to resolve airspeed issue - tools that computer doesn't have?
I would say 2 things - you can look out of the window (which is probably not gonna help) and you can use advanced processing algorithms (brain, or seat of the pants - whatever you want to call it).
One thing for sure - coumputer algorithms tend to improve. So does instrumentation.


Pitch and power settings. Becoming VMC is actually a big help, as is an ILS signal. Let's say the autopilot can go against everything it would normally do and determine to ditch its airspeed readings and even just fly according to pitch and power, how will it determine how/when to divert? At 30W is it really going to divert to KEF or YYT with 40 knot crosswinds and BLSN? How will it cope with a NAV AIR SATA SYS failure? Some decisions aren't binary and algorithms can't predict or determine the correct response in a cascading failure scenario.

And are you sure you decision on KEF vs YYT is the best one? (nothing personal, "you" = human pilot). If you made it to the runway, it was probably a good one anyway. The difference in terms of risk is probably not that major if you can fly another 2 hours to diversion point. Yes, regulations require minimal flight time and all that, but how big it actually is? You wouldn't hesitate between 1 and 3 hours flight time, it may be 95 vs 90 minutes where you would think twice. As for the flight path taking wind into account - computer may have a wind map, and I would certainly do a fine job calculating the times.

As for the best option... Did you ever had to look back at your own actions and say "well, I would be much better off doing things differently.." ? I certainly did, both personally and professionally.

Moral of the story - there are situations where human brain is better than machine, but those become fewer as time goes by. I wouldn't look for examples of human superiority in situations covered by checklists. It is unknown unknowns when humans are better. And it turns out that these are the situations where human can screw up.


This isn't about human superiority vs. checklist use by a computer - it's about a situation that a computer cannot diagnose because of its multi-faceted complexity.

As for the KEF/YYT situation that's why we use a decision making tool - TDODAR. How well can a computer generate options, make a decision and, as you refer to, review your decision. That's what the R is - review, we do it after taking the decision and its a continuous process all the way to a safe outcome. How well can a computer change its decision based upon new, previously unknown information?

Maybe we end up with the remote pilot drone system decades down the line but then you have a bunch of issues around interference etc.
 
prebennorholm
Posts: 7077
Joined: Tue Mar 21, 2000 6:25 am

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 11:00 pm

TVNWZ wrote:
The Air Force is already precision flying every day with nobody in the craft.

Correct. And at a high cost. It takes a lot more highly skilled people to control a pilotless aircraft.

All those people are a lot more expensive, even if they are working on the ground, if they are not sitting on an E-4 or such.

Some people seem to imagine that pilotless flight can save money due to reduced labor. Wrong. It is a lot more expensive and labor intensive.
Always keep your number of landings equal to your number of take-offs
 
kalvado
Posts: 2845
Joined: Wed Mar 01, 2006 4:29 am

Re: Airbus completes testing of self flying jet

Wed Jul 29, 2020 11:21 pm

BA777FO wrote:
kalvado wrote:
BA777FO wrote:

Pitch and power settings. Becoming VMC is actually a big help, as is an ILS signal. Let's say the autopilot can go against everything it would normally do and determine to ditch its airspeed readings and even just fly according to pitch and power, how will it determine how/when to divert? At 30W is it really going to divert to KEF or YYT with 40 knot crosswinds and BLSN? How will it cope with a NAV AIR SATA SYS failure? Some decisions aren't binary and algorithms can't predict or determine the correct response in a cascading failure scenario.

And are you sure you decision on KEF vs YYT is the best one? (nothing personal, "you" = human pilot). If you made it to the runway, it was probably a good one anyway. The difference in terms of risk is probably not that major if you can fly another 2 hours to diversion point. Yes, regulations require minimal flight time and all that, but how big it actually is? You wouldn't hesitate between 1 and 3 hours flight time, it may be 95 vs 90 minutes where you would think twice. As for the flight path taking wind into account - computer may have a wind map, and I would certainly do a fine job calculating the times.

As for the best option... Did you ever had to look back at your own actions and say "well, I would be much better off doing things differently.." ? I certainly did, both personally and professionally.

Moral of the story - there are situations where human brain is better than machine, but those become fewer as time goes by. I wouldn't look for examples of human superiority in situations covered by checklists. It is unknown unknowns when humans are better. And it turns out that these are the situations where human can screw up.


This isn't about human superiority vs. checklist use by a computer - it's about a situation that a computer cannot diagnose because of its multi-faceted complexity.

As for the KEF/YYT situation that's why we use a decision making tool - TDODAR. How well can a computer generate options, make a decision and, as you refer to, review your decision. That's what the R is - review, we do it after taking the decision and its a continuous process all the way to a safe outcome. How well can a computer change its decision based upon new, previously unknown information?

Maybe we end up with the remote pilot drone system decades down the line but then you have a bunch of issues around interference etc.

You're talking about multiple-choice decisions based on a closed set of information (divert- YES/NO; if YES - which of 3 nearby airports best fits the scope- based on distance, weather, availability of medical facility, ARFF) and effectively an algorithm for taking such decision. This is type of task where a computer can - and eventually will - excel humans. Feeding new information, if any, in the same algorithm is basically the review portion.

Basically I would say if you can envision something being a problem - it can be programmed along with resolution measures. It is situations of "who could have thought of it??" type that would matter.
 
User avatar
MillwallSean
Posts: 973
Joined: Tue Apr 08, 2008 3:07 pm

Re: Airbus completes testing of self flying jet

Thu Jul 30, 2020 12:18 am

Am I missing something here? we are arguing that no technological progress should happen and that two person flying aircrafts should, forever, be the norm. Hmm thats not how the world works. 30 years ago there were 3 persons flying airplanes. 50 years ago there were a team.
10 years from now it will be one person and the backup will sit in a boring suburban office, stepping in should something happen to the pilot (see dead-hand system on bullet trains etc).
I can guarantee that the flying public wont care whether there is 1 or 2 persons flying the plane just as they didn't care when we went from 3 to 2 persons. They will however care when we take the step to fully automated planes without pilots. That is a huge step, most of us humans wants safety / redundancy and we want to know that there is a human that can step in and think, not just analyse. But we don't need two humans to think, one is perfectly fine. And cockpits can be designed for one person these days, the tools, the help is more than satisfactory.

For me this step isn't so much a technological one, its more an institutional one. Unions etc will fight tooth and nail against it, trying their best to drum up support for maintaining status quo. Its a lost battle but I am sure that it will delay the implementation in certain jurisdictions. But for me, bring it on as humans we need to continuously evolve and improve and one person cockpits is a given on at least shorthauls.
No One Likes Us - We Dont Care.
 
BA777FO
Posts: 577
Joined: Sun Apr 22, 2018 2:58 pm

Re: Airbus completes testing of self flying jet

Thu Jul 30, 2020 5:01 am

FluidFlow wrote:

Look I respect what you are doing but all you do is defend your job on a 777, an archaic aircraft that needs a pilot to function.


I'm not defending my position - there are just some subjective situations that computers cannot determine the most appropriate course of action due to cascading failures or are unable to diagnose/recognise the problem. You mention GPS for speed - that'd be groundspeed, not airspeed, quite a key difference.

You talk about a bunch of sensors that can all be used - at the moment we can fly with deferred defects - the computerised aircraft won't be able to. That's a lot of extra maintenance and cancelled flights. The example you used about shutting down airspace around volcanoes will shut aviation down for weeks when we're now able to operate safely much closer to active areas.

It's nice that you think all of these issues can be overcome, maybe it will, but like I said, today's most advanced technology is just being delivered with a 30 year lifespan and the most advanced that has moved on is auto TCAS. You have a long way to go before it can make complex decisions.
 
IADFCO
Posts: 190
Joined: Sun May 22, 2016 4:20 pm

Re: Airbus completes testing of self flying jet

Thu Jul 30, 2020 5:31 am

@BA777FO -- I fully agree with your comments (coming from perhaps a different, R&D-heavy, world).

Autonomous flight in reasonably predictable situations is not easy, but is doable assuming (big assumption) enough computational power, enough sensors of the right kind, and enough reliability. The issue is unpredictable situations. Yes, for any given scenario one can conceive solutions that would address that particular scenario. The problem is that you have to conceive all possible scenarios, all possible ways in which the holes in the cheese can line up.

There isn't a magic "AI DVD" or "AI USB stick" that you plug in and automagically solves all problems. AI is very much garbage in-garbage out. You need to provide the rules and you need to train it. If you do it through machine learning you need to provide a ton of data, either through actual flight or at least through simulation, which however has to be so accurate that it can safely replace a real aircraft. Doable? maybe, and in a lab or carefully controlled conditions. Real life? Good luck. And this in a world where a major manufacturer cannot or does not want to account for a failed angle of attack sensor, and takes a year and a half (and counting) to fix it.

However, even if safe autonomous flight of a commercial airliner is probably a pipe dream for several decades to come, I'm absolutely convinced that R&D must continue. If 100 is not practical for the next 50 years, 10 or 20 can be reasonably achieved and still bring enormous benefits (e.g., reduced pilot workload), and be a stepping stone for 30 and 40, and so on.

Popular Searches On Airliners.net

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

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

Classic Airliners Props and jets from the good old days

Flight Decks Views from inside the cockpit

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

Cargo Aircraft Pictures of great freighter aircraft

Government Aircraft Aircraft flying government officials

Helicopters Our large helicopter section. Both military and civil versions

Blimps / Airships Everything from the Goodyear blimp to the Zeppelin

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

Accidents Accident, incident and crash related photos

Air to Air Photos taken by airborne photographers of airborne aircraft

Special Paint Schemes Aircraft painted in beautiful and original liveries

Airport Overviews Airport overviews from the air or ground

Tails and Winglets Tail and Winglet closeups with beautiful airline logos