Automation dependency stripped of political correctness.
Two points I would make. One, according to professor Winfield, who specialises in artificial intelligence, we're as far away from human level congnative thinking in computers as we are from faster than light space travel.
Two, why would you want autonomous airliners? Saftey? Or cost? As for Saftey, it's mathamaticly proven that if something can happen, it will. No matter how much redundancy you build into a system it can and will fail. Cost? The cost of the flight crew is about a pint of beer on every fight ticket. Hardly going to be noticed by most people.
So the question I ask is what's the point? Fully automated, autonomous airliners will still crash, will still kill people. The reduction in cost will be quickly reversed by the added cost of system redundancy. Which will also fail.
And no, Tourist, I didn't read your 250+ posts on the topic. I'm sure they are very well researched, and you obviously have a passion for a full automated airliner of the future. However the overall argument of why you would want to go down that route has not been made. They will still crash, and the cost savings are negligible, if any.
Automation dependency is a very real and dangerous threat. How you go about mitigating that threat is what we should be taking about. Not immersing ourselves in a far off fantasy.
Two, why would you want autonomous airliners? Saftey? Or cost? As for Saftey, it's mathamaticly proven that if something can happen, it will. No matter how much redundancy you build into a system it can and will fail. Cost? The cost of the flight crew is about a pint of beer on every fight ticket. Hardly going to be noticed by most people.
So the question I ask is what's the point? Fully automated, autonomous airliners will still crash, will still kill people. The reduction in cost will be quickly reversed by the added cost of system redundancy. Which will also fail.
And no, Tourist, I didn't read your 250+ posts on the topic. I'm sure they are very well researched, and you obviously have a passion for a full automated airliner of the future. However the overall argument of why you would want to go down that route has not been made. They will still crash, and the cost savings are negligible, if any.
Automation dependency is a very real and dangerous threat. How you go about mitigating that threat is what we should be taking about. Not immersing ourselves in a far off fantasy.
Join Date: Jan 2005
Location: W of 30W
Posts: 1,916
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by Tourist
The fact that an Airbus sometimes/usually saves you from yourself and when in direct law does not is hardly an indictment of the entire system.
Either way, if you hold the stick back you stay stalled. I'm glad you find the necessity to use both hands in the Boeing reassuring.
OK465,
We turned ADR 2 and 3 off to get them into ALTN law.
We turned ADR 2 and 3 off to get them into ALTN law.
Join Date: Jun 2007
Location: Wanderlust
Posts: 3,403
Likes: 0
Received 0 Likes
on
0 Posts
Stone_cold
Out of the speed brake inhibition conditions mentioned below in DSC-27-10-20 P 11/20 D only one is applicable in alternate law i.e. thrust lever above MCT. So just by dropping speed below Vls they will not retract unless thrust levers are advanced. Stall recovery procedure asks you to check they are retracted.
Speed brake extension is inhibited, if:
-SEC1 and SEC3 both have faults, or
-An elevator (L or R) has a fault, or
-Angle-of-attack protection is active, or
-Flaps are in configuration FULL, or
-Thrust levers above MCT position, or
-Alpha Floor activation.
If an inhibition occurs when the speed brakes are extended, they automatically retract and remain retracted until the inhibition condition disappears and the pilots reset the lever. (The speed brakes can be extended again, 10 s or more after the lever is reset).
Out of the speed brake inhibition conditions mentioned below in DSC-27-10-20 P 11/20 D only one is applicable in alternate law i.e. thrust lever above MCT. So just by dropping speed below Vls they will not retract unless thrust levers are advanced. Stall recovery procedure asks you to check they are retracted.
Speed brake extension is inhibited, if:
-SEC1 and SEC3 both have faults, or
-An elevator (L or R) has a fault, or
-Angle-of-attack protection is active, or
-Flaps are in configuration FULL, or
-Thrust levers above MCT position, or
-Alpha Floor activation.
If an inhibition occurs when the speed brakes are extended, they automatically retract and remain retracted until the inhibition condition disappears and the pilots reset the lever. (The speed brakes can be extended again, 10 s or more after the lever is reset).
Join Date: Feb 2004
Location: Sandpit
Posts: 177
Likes: 0
Received 0 Likes
on
0 Posts
Vilas ,
I am quite aware of the FCOM reference , hence my request . Your statement , without any qualification , said that the speedbrakes don't retract in alternate law . As you now state there is one case where the selecting TOGA( or above MCT ) will retract the speedbrakes , regardless of the current law .
Unreliable speed and GPWS recoveries also require one to check speedbrake retracted .
I am quite aware of the FCOM reference , hence my request . Your statement , without any qualification , said that the speedbrakes don't retract in alternate law . As you now state there is one case where the selecting TOGA( or above MCT ) will retract the speedbrakes , regardless of the current law .
Unreliable speed and GPWS recoveries also require one to check speedbrake retracted .
Join Date: Jun 2007
Location: Wanderlust
Posts: 3,403
Likes: 0
Received 0 Likes
on
0 Posts
Vilas ,
I am quite aware of the FCOM reference , hence my request . Your statement , without any qualification , said that the speedbrakes don't retract in alternate law . As you now state there is one case where the selecting TOGA( or above MCT ) will retract the speedbrakes , regardless of the current law .
Unreliable speed and GPWS recoveries also require one to check speedbrake retracted .
I am quite aware of the FCOM reference , hence my request . Your statement , without any qualification , said that the speedbrakes don't retract in alternate law . As you now state there is one case where the selecting TOGA( or above MCT ) will retract the speedbrakes , regardless of the current law .
Unreliable speed and GPWS recoveries also require one to check speedbrake retracted .
Join Date: Jun 2000
Location: last time I looked I was still here.
Posts: 4,507
Likes: 0
Received 0 Likes
on
0 Posts
Have FBW a/c become too complicated for their own good? There is much talk about the pilot being detached from the a/c in modern automation, and hence the concern about automation dependancy. This discussion has focused on improving manual skills to counter this. In these latest threads it seems the detachment from the a/c could be because of over complication about what it is going to do and when. Surely it is 'too much' to have a pilot in a non-normal situation, with all the associated stresses, where the pilot is unsure of what flight control system will do what and with what limitations. Surely, in a non-normal scenario things should be as simple and as basic as possible.
I admit I am not an AB pilot. I apologise if I've missed something, but there does seem an unresolved debate amongst the AB crews about what is correct in these various scenarios.
Is that healthy, or have the design engineers, inadvertently, squeezed the pilot out of the loop at a time when they need to be 'on top of it all'? If that is true, my concern is it will get worse before it gets better. What is on the drawing board now that we, the pilots, have no idea about? I'm not sure manufacturer's test pilots are the best to judge what is the best design for the average line pilot.
I admit I am not an AB pilot. I apologise if I've missed something, but there does seem an unresolved debate amongst the AB crews about what is correct in these various scenarios.
Is that healthy, or have the design engineers, inadvertently, squeezed the pilot out of the loop at a time when they need to be 'on top of it all'? If that is true, my concern is it will get worse before it gets better. What is on the drawing board now that we, the pilots, have no idea about? I'm not sure manufacturer's test pilots are the best to judge what is the best design for the average line pilot.
Join Date: Apr 2008
Location: Nearer home than before!
Posts: 524
Likes: 0
Received 0 Likes
on
0 Posts
Totally agree RAT5.
The thing about a Boeing is you can see clearly what the other guy is doing because your yolk moves too. Whatever mode it is in it will appear to behave like an airplane, pure and simple. Predictable and simple. Many of us like this.
KISS is a powerful safety aid.
The differing modes and laws in an AB complicate at a time when you need to keep it simple. This is undeniable. What's it doing now, what law is it in? Will it help or not? Discussions like this at critical edge of envelope times are not ideal.....
The thing about a Boeing is you can see clearly what the other guy is doing because your yolk moves too. Whatever mode it is in it will appear to behave like an airplane, pure and simple. Predictable and simple. Many of us like this.
KISS is a powerful safety aid.
The differing modes and laws in an AB complicate at a time when you need to keep it simple. This is undeniable. What's it doing now, what law is it in? Will it help or not? Discussions like this at critical edge of envelope times are not ideal.....
Vilas,
We were simulating at stall in level flight at 5000 feet. The levers were at idle. Auto trim worked until well into VLS but stopped well before Vsw (Alpha SW) on both the 380 and 330. The speed brakes auto retracted once we entered VLS in both the 380 and the 330. I don't know how else to explain it.
As I said, I have no clue if the mini buses behave differently.
We were simulating at stall in level flight at 5000 feet. The levers were at idle. Auto trim worked until well into VLS but stopped well before Vsw (Alpha SW) on both the 380 and 330. The speed brakes auto retracted once we entered VLS in both the 380 and the 330. I don't know how else to explain it.
As I said, I have no clue if the mini buses behave differently.
RVF750,
Seeing the yoke move did not save Turkish (737 stall), Flash airlines (737 loss of control) nor Ethiopian (737 loss of control).
When the pilot is not up to the job, it does matter how simple it is.
The bus is just another aeroplane. But with better seats.
Seeing the yoke move did not save Turkish (737 stall), Flash airlines (737 loss of control) nor Ethiopian (737 loss of control).
When the pilot is not up to the job, it does matter how simple it is.
The bus is just another aeroplane. But with better seats.
Join Date: Jan 2001
Location: Home
Posts: 3,399
Likes: 0
Received 0 Likes
on
0 Posts
I think that it is probably correct to say that an Airbus is more tricky for the pilots once out of normal law than a Boeing is.
This is reasonable, because suddenly not only do you have a problem, but your aircraft suddenly changed a bit and all your natural responses are not quite right any more.
That is not the whole story though, is it.
Accident statistics show that Boeing and Airbus are each as safe as each other, which can only mean that whilst the systems in an Airbus are actually working, the Airbus is safer to offset the Boeing advantage once non-normal.
That would suggest that if you can make the automatics more reliable, the Airbus wold be the safest?
Since the Airbus systems are frankly archaic, that should not be too hard.
This is reasonable, because suddenly not only do you have a problem, but your aircraft suddenly changed a bit and all your natural responses are not quite right any more.
That is not the whole story though, is it.
Accident statistics show that Boeing and Airbus are each as safe as each other, which can only mean that whilst the systems in an Airbus are actually working, the Airbus is safer to offset the Boeing advantage once non-normal.
That would suggest that if you can make the automatics more reliable, the Airbus wold be the safest?
Since the Airbus systems are frankly archaic, that should not be too hard.
AF447 has been, and continues to be debated with polarising views between man and machine.
Many posts reflect the difficulties in avoiding hindsight bias and the need for systems thinking, whereby complex interactions in accidents appear simpler to explain via individual contributors, and thus each becomes a basis for a solution. Often there is focus on the actual event, overlooking the hidden, latent precursors.
Life, aviation, and accidents are rarely simple. Accidents involve many contributing factors, each necessary, but where none in isolation may be sufficient to cause it.
A bold reversal of the latter is that no one factor will be assured of preventing future accident with the degree of certainty often stated.
We have a natural dislike of uncertainty and complexity, which can lead to inaccurate simplifications. Modern safety initiatives seek to manage this; the need to change the way we think about safety, the role of the human, what is error, operational expectations vs actually, and the ability to learn from accidents, incidents and events.
Such a change requires us to accept that nothing is certain, and how individually and collectively we might manage uncertainty. Rarely can we identify a single dominating (proven) factor in accidents; alternatively groups or patterns of contributions might indicate areas of interest according to viewpoint.
Safety improvements reside in what can be learnt and applied; a reactive start, seeking proactive improvement. This requires all management levels to be involved, regulator, operator, pilot.
Everyone should have something to learn - what if, why. Questions are easy to think of, but not so easy pose in context, with prior consideration of what is meaningful in operations. Also, a bottom up approach, where pilots question operators, operators question regulators, may be time consuming, and against the flow of safety management.
Alternatively, a top down approach could provide greater benefit where the regulator / operator consider what is ‘meaningful’ according the front line actors.
Safety in an industry often reflects its management – it regulation.
The thought examples in http://www.pprune.org/9242725-post116.html #116 do not seek agreement or otherwise – contributors or solutions – each depends on viewpoint, context, mind-set, but if they do relate to current operations then they could be the basis of safety improvement.
Each of us needs to challenge and provoke our beliefs, where these thoughts might collectively (all of us) provide some insight of what is required. The outcome might only require small, well-reasoned and affordable changes, spanning many subjects, but remembering that uncertainty reigns, it necessitates judgement.
What might we learn?
Many posts reflect the difficulties in avoiding hindsight bias and the need for systems thinking, whereby complex interactions in accidents appear simpler to explain via individual contributors, and thus each becomes a basis for a solution. Often there is focus on the actual event, overlooking the hidden, latent precursors.
Life, aviation, and accidents are rarely simple. Accidents involve many contributing factors, each necessary, but where none in isolation may be sufficient to cause it.
A bold reversal of the latter is that no one factor will be assured of preventing future accident with the degree of certainty often stated.
We have a natural dislike of uncertainty and complexity, which can lead to inaccurate simplifications. Modern safety initiatives seek to manage this; the need to change the way we think about safety, the role of the human, what is error, operational expectations vs actually, and the ability to learn from accidents, incidents and events.
Such a change requires us to accept that nothing is certain, and how individually and collectively we might manage uncertainty. Rarely can we identify a single dominating (proven) factor in accidents; alternatively groups or patterns of contributions might indicate areas of interest according to viewpoint.
Safety improvements reside in what can be learnt and applied; a reactive start, seeking proactive improvement. This requires all management levels to be involved, regulator, operator, pilot.
Everyone should have something to learn - what if, why. Questions are easy to think of, but not so easy pose in context, with prior consideration of what is meaningful in operations. Also, a bottom up approach, where pilots question operators, operators question regulators, may be time consuming, and against the flow of safety management.
Alternatively, a top down approach could provide greater benefit where the regulator / operator consider what is ‘meaningful’ according the front line actors.
Safety in an industry often reflects its management – it regulation.
The thought examples in http://www.pprune.org/9242725-post116.html #116 do not seek agreement or otherwise – contributors or solutions – each depends on viewpoint, context, mind-set, but if they do relate to current operations then they could be the basis of safety improvement.
Each of us needs to challenge and provoke our beliefs, where these thoughts might collectively (all of us) provide some insight of what is required. The outcome might only require small, well-reasoned and affordable changes, spanning many subjects, but remembering that uncertainty reigns, it necessitates judgement.
What might we learn?
There are numerous incidents where bus pilots have made dual inputs.
Yes there is the procedure to use sidestick priority until full handover of controls is complete, but in the heat of the moment, particularly if not particularly experienced on Airbus, it's not the most naturally intuitive.
With all the magic in an Airbus I don't see why they couldn't have linked controls so you can feel or see what is happening to the stick and so would prevent the control conflict.
Join Date: Jun 2007
Location: Wanderlust
Posts: 3,403
Likes: 0
Received 0 Likes
on
0 Posts
donpizmeov
I am not trying to prove you wrong but I am trying to understand the protection. Speed brakes will retract if low speed stability is available but I am talking about alternate2 with low speed stability not available. I am quoting below from A330 FCOM which says same thing as A320:
SPEEDBRAKE CONTROL
The pilot controls the speed brakes with the speed brake control lever. The speed brakes involve
spoilers 1 to 6. Speed brake extension is inhibited, if:
‐ Maneuver Load Alleviation (MLA) is activated
‐ Angle-of-attack protection is active
‐ Low speed stability is active
‐ At least one thrust lever above MCT
‐ Alpha floor is activated.
If an inhibition occurs when the speed brakes are extended, they automatically retract and stay retracted until the inhibition condition disappears, and the pilot resets the lever.
I am not trying to prove you wrong but I am trying to understand the protection. Speed brakes will retract if low speed stability is available but I am talking about alternate2 with low speed stability not available. I am quoting below from A330 FCOM which says same thing as A320:
SPEEDBRAKE CONTROL
The pilot controls the speed brakes with the speed brake control lever. The speed brakes involve
spoilers 1 to 6. Speed brake extension is inhibited, if:
‐ Maneuver Load Alleviation (MLA) is activated
‐ Angle-of-attack protection is active
‐ Low speed stability is active
‐ At least one thrust lever above MCT
‐ Alpha floor is activated.
If an inhibition occurs when the speed brakes are extended, they automatically retract and stay retracted until the inhibition condition disappears, and the pilot resets the lever.
Last edited by vilas; 3rd Feb 2016 at 14:19.
This was demonstrated to me when I did my command a bit over a decade ago. Still works now.
When you apply speed brake you increase VLS and Vsw (alpha SW). By lowering the speed brake you get more lift quicker than taking flap. If they didn't auto retract, the getting rid of the speed brakes would be one of the first actions of the stall recovery. But it's not even considered until you are out of the stall and have added thrust. Have you ever why?
The point being also, that if you need to hold pressure on the stick to maintain attitude in ALTN law, you are at very low speed and should be doing something about it.
When you apply speed brake you increase VLS and Vsw (alpha SW). By lowering the speed brake you get more lift quicker than taking flap. If they didn't auto retract, the getting rid of the speed brakes would be one of the first actions of the stall recovery. But it's not even considered until you are out of the stall and have added thrust. Have you ever why?
The point being also, that if you need to hold pressure on the stick to maintain attitude in ALTN law, you are at very low speed and should be doing something about it.
Join Date: Feb 2015
Location: Alternate places
Age: 76
Posts: 97
Likes: 0
Received 0 Likes
on
0 Posts
FBW of course is just another way of moving flight controls, with the additional requirement of feedback and the quality of "gain".
"Protections" are an entirely different matter, made possible by digital flight control. FBW and protections aren't separate but FBW itself (C*) is not "protection".
In the failure cases where various inputs for protections are no longer valid there has to be a way of maintaining pilot control over the aircraft.
This may seem like stating the obvious and perhaps it is, but sometimes the two can be conflated which can lead to incorrect conclusions. FBW itself, will still stall an aircraft just like any other design. How graceful the reversion is, is a matter of design, and obviously opinion! I can say that in the sim, such reversions are non-events.
The Airbus is a complicated aircraft and so is the B777, more particularly, the B787 which is FBW with protections. All this we know. The "nodes" of the discussion regarding the Airbus revolved around AF447, specifically Alternate Law behaviour, (not around the AirAsia accident however; pulling flight control CBs in-flight is strictly forbidden). The point has been made many times, that thirty-one other crews encountered a UAS event and wrote it up in the logbook. It simply isn't/wasn't an emergency. The ensuing at-length discussion generated by AF447 has however, highlighted the Airbus, in particular those extremely rare circumstances which, like other designs, occur as anticipated, with a rarity approaching 10^-9 or so. The millions of hours of successful, unremarkable flight, though silent, perhaps tells us that concern over autoflight systems (and the 'unresolved debate' matter), may be disproportionate to the reality. Everyone just gets on with flying the airplane they're on, A or B.
In response to your question/observation regarding unresolved debates, part of this is due to not knowing the aircraft well enough or at all, part of it has to do with mythology, prejudice & politics and part has to do with a genuine concern by those who really know their stuff and still have questions regarding some design features, (two of which have been mentioned in the thread - THS movement continuing during stall & silent stall warning with NCD). These aircraft, (present Boeing and Airbus) are all far too complex for any one person or the pilots themselves, to absorb and retain. Normal flight operations occurs 99%+ of the time for all types.
The statistical record, maintained by Boeing since 1959, speaks to the question regarding complexity and competency - neither design (A/B) are "way out in front". However, the A319/A320/A321 series' hull-loss-fatal-accident rates is slightly lower that the equivalent-type, B737 rates, (per million departures).
Last edited by FDMII; 4th Feb 2016 at 01:30. Reason: Fix link to Boeing doc
Join Date: Sep 2010
Location: Angels 20 and climbing
Posts: 81
Likes: 0
Received 0 Likes
on
0 Posts
A very interesting debate here on automation/autonomy from a Royal Aeronautical Society 150th anniversary event last month...
"This House believes there will be no need for pilots 40 years from now”
Royal Aeronautical Society | Podcast | Black Tie Debate
Some great speakers and stimulating discussion...
"This House believes there will be no need for pilots 40 years from now”
Royal Aeronautical Society | Podcast | Black Tie Debate
Some great speakers and stimulating discussion...