For those like me who, though not a pilot, enjoy watching Aircrash Investigation and similar programs, and finds the investigations and findings inspiring/sobering/educational as a software engineer, and like me weren't aware of the "Children of [the] Magenta" phrase, this article and audio piece [1] looks a good place to start. The phrase was used in a 1997 presentation - available publicly [2] - by American Airlines captain Warren Van Der Burgh, who said that (quoting the article's summary of his words): "The industry has turned pilots into “Children of the Magenta” who are too dependent on the guiding magenta-colored lines on their screens." He makes his case with some very funny lines, too. Quite the humorist: "I got stuff to worry about, don't I. Heads up stuff. It's probably not a good time to be typing." [3] immediately precedes his explanation of the phrase.
It's an area of ongoing research as well, e.g. the Humans and Autonomy Lab at Duke. (Missy Cummings, who is the director of the lab, used to be a Navy fighter pilot.)
I expect we'll hear a lot more on this topic. Assuming you're at all skeptical about the timeframe for full automobile driving automation, it seems a certainty that we'll have a (significant) period when stuff works most of the time--until it doesn't. And, given that lots of people drive distracted today, what's it going to look like when the car can drive down the highway most of the time on its own?
Isn't this a false dichotomy, e.g., shouldn't it be "The problem is relying on automation in addition to real piloting skills"?
Look at the space program, for example. Extremes of forces and very short windows for decision-making have moved spaceflight in the direction of automation for most mission critical moments. The data that feeds in-flight automation is the same that feeds simulators, so I am not sure where the difference lies in your differentiation between the two.
I can see a point where doing everything by software rather than human is safer and makes the most sense (see the direction Tesla is taking us), and at that point there will be no issue. But that is not the case now.
> The data that feeds in-flight automation is the same that feeds simulators
The problem is when the data sources fail, like in AF447.
Or when you have people crashing an airplane in perfect weather, like Asiana Airlines 214 in San Francisco
Yes, automation makes things easier, but they fail eventually. And when it fails you have to fly it manually (or with a lower level of automation). But some modern pilots are incapable of doing that.
Well, I am not arguing for or against automation (although ultimately as an optimistic programmer I land very much in the for camp.)
What I am saying is that simulation time alone is not good enough. Nor is flying-only time. Both are needed, and for different reasons. Sure, time spent in automation mode in the air is not incredibly useful. But it is reality, and enduring that, only to have to take manual control in a panic after 8 hours of boredom is something I don't see simulators re-reproducing very well any time soon.
You don't put the pilot in an 8h training flight. You train for specific failures in different situations in a simulator (take off/cruise/landing). And you're right, there's no training for "doing nothing" for 8h and then having an issue
Level of automation you're using has nothing to do with flying on a simulator or an actual plane.
Simulators are great for training on emergency scenarios that can't easily or safely be done in a real airplane, or practicing and learning new avionics or systems but they shouldn't be a substitute for real flying experience. This is a bad decision.
Completely putting aside the question of the quality of the simulation of the aircraft and physics involved, one possible reason is that simulations can remove extreme emotions from the equation. The pilot knows they won't die in the simulator if they make an error. (This can be both a positive and negative gain.)
To take it to the extreme, would you be willing to attempt landing a real 747 solo, having only ever flown using simulators?
I will be willing to do it. I won't be willing to be present in an aircraft while someone elae does it. Not because I am a pilot - just I am ok with dying from my own mistake.
"I was just settling down with my book and I felt the airplane do a very odd maneuver. We could feel the airplane doing something very significant and abnormal."[1][2]
Human brains are amazing things that, with no special effort - only practice and repetition - can combine a huge number of disparate inputs to synthesize the "feel" of something.
The airplane is a physical thing and it has "a feel". A good pilot (or jockey or surgeon or fighter or truck driver) has a tremendous amount of stored information and experience related to "the feel" of what they are doing. We would be vastly worse off (and less safe) if we discounted that human ability and fooled ourselves into believing that simulation can ever match the experience of doing a thing in the physical world.
No simulator can recreate the forces experienced in a real plane exactly, and it can be very disorienting and nauseating. As a simple example, consider a plane accelerating down the runway for takeoff --- that feeling of thrust that pushes you back into the seat can't be simulated. I have severe motion sickness and can't do things like play FPS, simulators, or watch videos with lots of movement in my field of vision without becoming very dizzy and fatigued. VR and 3D only makes it worse. Yet I have no problems driving or riding in a real plane.
I built one of these certified simulators; they are not what you will find at most people's desks. We built a cave with a large number of 120Hz 1080 projectors custom modified to sync in tandem to shutter glasses, and used a medical device for sub-millimeter head position tracking. The cockpit was an actual, decommissioned helicopter with working interfaces.
It was an incredible experience to build and operate; I never did stick around for the final install, which was going to mount the cockpit onto a platform capable of rotating and lifting in a variety of crazy ways.
What is the purpose of the head position tracking?
Did the aircraft you build the simulator for has some sort of advanced HUD that always shows correctly no matter the position of the pilot head??
So it would feel real; you could look out the side window, or shift your weight to look around trees or to get a better view down the side of the helicopter.
I work in VR now and honestly nothing comes close to the experience we built.
>I work in VR now and honestly nothing comes close to the experience we built.
This is an interesting statement, because in all the wide sphere of points where VR can be applied, it seems to me that the only place of any 'validity' or .. true social value .. (i.e. not gaming/entertainment) .. is in the realm of training for high-performance/life-critical operator scenarios.
Which is not to say that these things can/can not be gamified - but perhaps that the end result of such a stringent effort by society to apply VR technology to training and simulation of otherwise-life-threatening roles/responsibilities/duties is that it all becomes a bit of a game.
What I mean to say is, why is there such a gap between "work in VR" and "built serious simulator system", in your opinion?
> in all the wide sphere of points where VR can be applied, it seems to me that the only place of any 'validity'... is in the realm of training for high-performance/life-critical operator scenarios.
We're already seeing socially important applications outside of training. VR is being used as part of treatment for PTSD and phobia; it can provide a very believable experience of a traumatic stimulus in a safe environment. VR may prove to be a powerful tool for psychotherapists.
More broadly, I think we have underestimated the psychological significance of presence. Users respond to convincing VR experiences in a totally different way to other media. This difference seems to be relatively durable and does not fully diminish when the novelty wears off. Pilots in simulators experience real physiological stress, even when they are experienced simulator users.
How would the world change if most of us got our news through VR - if we felt that we were stood in Mosul or Aleppo, flinching at every bang, reflexively crouching down for safety?
>What I mean to say is, why is there such a gap between "work in VR" and "built serious simulator system", in your opinion?
There's a vast difference between a $2000 consumer system and a $2,000,000 simulator. A headset can't approach the fidelity of a real cockpit with real controls and instruments, mounted on a huge hydraulic motion rig. VR technology is constantly advancing, but there will always be a gap between an expensive single-purpose system and a cheap general-purpose system.
Also, when using shutter glasses the best projection is possible when you know the exact position of the viewer. The sense of depth and presence was unlike anything I had experienced; the Vive is sort of a cludgy close experience, but the visual fidelity is nowhere near what we made.
It would be nice if we had either public driving simulators or classes that you could use to improve your driving and response to erratic other drivers. I suspect that the lack of this deliberate training is one reason why driving is much more dangerous than flying. Back when I owned a car, it seemed like all of the classes on driving marketed themselves as either:
1) A totally unchallenging formality that will satisfy the requirements for keeping your license.
2) Training for police to use a car as tactically while being shot at.
There didn't seem to be any middle ground for people who just felt like they'd like to reduce their chances of killing a cyclist.
I wonder if it would profitable to build an arcade machine and charge people to play a more immersive version of Crazy Taxi, GTA, or some prohibition bootlegger simulator.
Fundamentally, I think most drivers view driving as just a thing they do, and give no thought to the possibility of improving. So having classes or simulators isn't going to help, because they wouldn't take advantage of them.
The thing is that it's not so much that most people don't know how they could drive more defensively/safer/etc., it's that they're in a hurry, or they're texting, or otherwise aren't carefully monitoring their surroundings.
That is an amazing idea. I would love it if people were required to spend time in driving simulators practicing spin recovery, emergency braking, etc. This is totally doable and could have a huge impact on accident statistics. It's also a much better way to validate drivers' reaction times and decision-making ability as they age.
In an ideal world, sure. In the real world, what could be the problem with making everyone take a regular test to have the right to continue doing something that's an inherent part of most people working and participating in society?
Yes, there are some people driving who probably shouldn't be on the road. But until such times as automation arrives, you'll find exactly zero support for that idea outside the young and fit of Silicon Valley.
I strongly disagree that driving is a right. It is a privilege that should be validated as well as is practical to protect yourself and the rest of society. I don't think we are anywhere near the limits of what is practical. Because we don't currently do this doesn't make it right, or something we shouldn't think of ways to solve. See recent changes in North American society regarding smoking laws. Change can happen faster than you think.
Doesn't matter if it's a birthright (and the use of "to have the right" in that sentence was more the meaning of "entitlement"). It's "an inherent part of most people working and participating in society". You need to change that if you want it to be reasonable to stop more than a few extreme cases from driving.
Exactly. When the day comes when there are 100% autonomous vehicles--which I don't expect to be particularly soon, but that's another discussion--I think it may be seen as [EDIT: wording] perfectly reasonable to impose stricter licensing requirements on manual vehicles operating on public streets.
I'd also add that fatalities are far more often caused by reckless young drivers than by the not so fit and young whose reflexes or eyesight may not be what they once were.
Most racing schools offer courses that are intended to improve street driving skill - e.g. winter driving is a common one, often with a "skid car" rigged with casters that remove much of the car's weight from the tires so that it will go into slides easily (as if it were on ice). There are also often "defensive driving" courses that cover swerving and fast braking under control. I've taken one of these courses and thought it was absolutely useful and improved my driving. That said, it was pretty expensive, and not well publicized.
Basic driver's education in the state I grew up in was state subsidized and offered through the community college so it was quite inexpensive. On the flipside, it was a terrible course, with minimal actual driving time and a terrible case of design for the lowest common denominator. I wonder if more advanced driving courses could be offered the same way, but perhaps they'd fall to the same problems.
You might look at motorcycling as a positive example. While many states require the Motorcycle Safety Foundation basic course for a motorcycle endorsement, MSF also publishes an advanced course and various skill courses that seem to be relatively popular amongst motorcyclists. MSF courses are developed by the Foundation and all the materials are licensed (I would assume at nominal cost) to various local organizations that actually teach them, in my area affiliated also with the community college.
As I alluded to in another comment, one of the issues here is that, for most adults in the US, driving is pretty much part and parcel of working and functioning in society. And wishing that everyone lived in cities and that we had futuristic mass transit systems or 100% autonomous cars doesn't change today's reality.
Therefore, if we're going to require advanced multi-day driving schools and/or annual physical driving exams (or whatever), you're going to have to explain how that works in a world where many people can barely afford to keep their clunker running and are unemployed if they can't drive.
Driving may not be a "right" (whatever that means exactly as we place limits on lots of rights), but I wouldn't want to be the politician who decides to take that "privilege" away from 20% of the population because they missed their annual refresher course or recertification exam.
The UK driving test's "hazard perception" is a hugely watered-down version of this - you watch a video and press a button when you think you've encountered a hazard.
I think a lot of people would benefit from being able to practice their parking in VR rather than with a >£25/h driving instructor.
In Norway to get a driver license one has to take one-way training where they simulate icy road with field covered with oil. I still remember the feeling when the instructor used hand brake and the car rotated 360 degrees. They also demonstrate how helpful all that modern anti-sliding equipment in cars by disabling it.
I was under the impression that 95% of the time, the pilot sits in the cockpit doing nothing, other that watching the autopilot by a clear sunny day. Whereas in a simulator the pilot is busy practicing challenging environments.
I am not qualified to have an opinion on whether a pilot can do simulators only but intuitively, an hour of simulator seems to be worth more than one hour of real life flight.
Sure, they are both valuable, and simulators will let you practice things you never want to happen in real life, so you can be prepared. The issue is that simulators are being treated as if they can completely substitute for real flying experience. E.g., the mere fact that you may be bored most of the time in a long flight can affect how you perform during the 5% where you really need to be present.
I suspect though that knowing you're being tested in a simulator, even if it is a boring long-endurance mission, could still affect how you think about it and approach it.
Is CAE a major player among flight sim companies? Because in that case, there's clear economic incentive in making simulator training accepted.. first in Canada, with the hope that other countries will follow. Big company will get a good chance to export their product.
It seems to me that the #1 way to predict pilot performance in emergencies is whether the pilot has encountered the emergency before and lived. If so, running the pilot through the most scenarios possible in a given unit time would seem to be a good strategy to train pilots, and simulators would increase the rate of scenarios per unit time versus a real aircraft.
However, you can't get away from actually flying a plane. There's little things you don't practice in the simulator, like making sure you have your handheld VHF radio, checking for water in the fuel tanks, and how to balance cargo, that you'll never learn in front of a computer screen.
If it's anything like network administration, I'd say something like 20% real experience vs 80% simulated might be a good mix. People who favor just one or the other lack performance under adverse circumstances compared to admins who do both, I've noticed.
Ah politics. It appears that this story is really about getting more funding for flight time for aviation inspectors.
From TFA:
> Many of Transport Canada’s aviation inspectors, some of whom are licensed pilots, are not able to keep their licences current because Conservative spending cuts imposed in 2011 curtailed their ability to fly government aircraft. He believes the department changed the rules to ensure federal aviation inspectors maintain the proper credentials.
> A CFPA survey of licensed aviation inspectors, conducted in April, found that about half have not been assigned flight training for more than a year and one in 10 has not been assigned to flight training for up to 10 years or more.
> “With the stroke of a pen, Transport Canada has devised a work-around so their own pilots comply with Canadian Aviation Regulations,” Mr. McConnell said. “Aviation inspectors who are pilots are becoming like traffic cops without a driver’s licence. This situation undermines public safety and inspectors’ credibility with the industry they are supposed to oversee.”
This article is factually incorrect. Here's some glaring examples:
> United States and European Union, which still compel pilots to meet stringent standards to keep their licences valid, including a requirement to actually fly a plane.
Not true, at least not in the US. According to 14 CFR 61.56 (i) "A flight simulator or flight training device may be used to meet the flight review requirements of this section"
> There is a whole host of skills you don’t exercise in a simulator
True, but the opposite is also true. I can think of number of scenarios that can be practiced in a sim but cannot be safely or legally practiced in aircraft. Example of such scenario: engine failure in instrument meteorological conditions.
> If you had a heart issue, would you go and see a heart surgeon that hadn’t operated in five years
I wonder if Mr. McConnell is aware that every single line pilot who files B787 or A380 today did all their training for that airframe and passed the rating test in a flight simulator.
The reality is all airline pilots train in flight sims these day. There is no practical way of doing that in actual aircraft. If one can do all their training and pass the license test in a sim why can't they do periodic flight review in a sim?
49 comments
[ 3.1 ms ] story [ 120 ms ] threadhttp://www.tc.gc.ca/media/documents/ca-publications/9-LRA-e....
http://www.rcinet.ca/en/2016/10/04/pilots-licence-simulator-...
I would love it if a developer involved in Transport Canada approved simulator software could comment.
The problem is relying on automation instead of real piloting skills
I trust much more a pilot practicing manual flying in a simulator than doing real flying in "Children of Magenta" mode
[1] http://99percentinvisible.org/episode/children-of-the-magent...
[2] https://youtu.be/pN41LvuSz10
[3] https://youtu.be/pN41LvuSz10?t=510
I expect we'll hear a lot more on this topic. Assuming you're at all skeptical about the timeframe for full automobile driving automation, it seems a certainty that we'll have a (significant) period when stuff works most of the time--until it doesn't. And, given that lots of people drive distracted today, what's it going to look like when the car can drive down the highway most of the time on its own?
Look at the space program, for example. Extremes of forces and very short windows for decision-making have moved spaceflight in the direction of automation for most mission critical moments. The data that feeds in-flight automation is the same that feeds simulators, so I am not sure where the difference lies in your differentiation between the two.
I can see a point where doing everything by software rather than human is safer and makes the most sense (see the direction Tesla is taking us), and at that point there will be no issue. But that is not the case now.
> The data that feeds in-flight automation is the same that feeds simulators
The problem is when the data sources fail, like in AF447.
Or when you have people crashing an airplane in perfect weather, like Asiana Airlines 214 in San Francisco
Yes, automation makes things easier, but they fail eventually. And when it fails you have to fly it manually (or with a lower level of automation). But some modern pilots are incapable of doing that.
What I am saying is that simulation time alone is not good enough. Nor is flying-only time. Both are needed, and for different reasons. Sure, time spent in automation mode in the air is not incredibly useful. But it is reality, and enduring that, only to have to take manual control in a panic after 8 hours of boredom is something I don't see simulators re-reproducing very well any time soon.
Level of automation you're using has nothing to do with flying on a simulator or an actual plane.
You can fly in a real aircraft for hours doing nothing but twisting autopilot knobs. OTOH you can practice basic airmanship skills in a sim.
EDIT: sorry, I realized that you were basically saying the same thing.
To take it to the extreme, would you be willing to attempt landing a real 747 solo, having only ever flown using simulators?
Human brains are amazing things that, with no special effort - only practice and repetition - can combine a huge number of disparate inputs to synthesize the "feel" of something.
The airplane is a physical thing and it has "a feel". A good pilot (or jockey or surgeon or fighter or truck driver) has a tremendous amount of stored information and experience related to "the feel" of what they are doing. We would be vastly worse off (and less safe) if we discounted that human ability and fooled ourselves into believing that simulation can ever match the experience of doing a thing in the physical world.
[1] http://jalopnik.com/5629528/how-i-saved-a-747-from-crashing
https://en.wikipedia.org/wiki/Simulator_sickness
No simulator can recreate the forces experienced in a real plane exactly, and it can be very disorienting and nauseating. As a simple example, consider a plane accelerating down the runway for takeoff --- that feeling of thrust that pushes you back into the seat can't be simulated. I have severe motion sickness and can't do things like play FPS, simulators, or watch videos with lots of movement in my field of vision without becoming very dizzy and fatigued. VR and 3D only makes it worse. Yet I have no problems driving or riding in a real plane.
It was an incredible experience to build and operate; I never did stick around for the final install, which was going to mount the cockpit onto a platform capable of rotating and lifting in a variety of crazy ways.
I work in VR now and honestly nothing comes close to the experience we built.
This is an interesting statement, because in all the wide sphere of points where VR can be applied, it seems to me that the only place of any 'validity' or .. true social value .. (i.e. not gaming/entertainment) .. is in the realm of training for high-performance/life-critical operator scenarios.
Which is not to say that these things can/can not be gamified - but perhaps that the end result of such a stringent effort by society to apply VR technology to training and simulation of otherwise-life-threatening roles/responsibilities/duties is that it all becomes a bit of a game.
What I mean to say is, why is there such a gap between "work in VR" and "built serious simulator system", in your opinion?
To answer your question, the difference is in fidelity. HMDs are slow, low resolution, heavy and uncomfortable compared to a properly built cave.
We're already seeing socially important applications outside of training. VR is being used as part of treatment for PTSD and phobia; it can provide a very believable experience of a traumatic stimulus in a safe environment. VR may prove to be a powerful tool for psychotherapists.
More broadly, I think we have underestimated the psychological significance of presence. Users respond to convincing VR experiences in a totally different way to other media. This difference seems to be relatively durable and does not fully diminish when the novelty wears off. Pilots in simulators experience real physiological stress, even when they are experienced simulator users.
How would the world change if most of us got our news through VR - if we felt that we were stood in Mosul or Aleppo, flinching at every bang, reflexively crouching down for safety?
>What I mean to say is, why is there such a gap between "work in VR" and "built serious simulator system", in your opinion?
There's a vast difference between a $2000 consumer system and a $2,000,000 simulator. A headset can't approach the fidelity of a real cockpit with real controls and instruments, mounted on a huge hydraulic motion rig. VR technology is constantly advancing, but there will always be a gap between an expensive single-purpose system and a cheap general-purpose system.
1) A totally unchallenging formality that will satisfy the requirements for keeping your license.
2) Training for police to use a car as tactically while being shot at.
There didn't seem to be any middle ground for people who just felt like they'd like to reduce their chances of killing a cyclist.
I wonder if it would profitable to build an arcade machine and charge people to play a more immersive version of Crazy Taxi, GTA, or some prohibition bootlegger simulator.
Fundamentally, I think most drivers view driving as just a thing they do, and give no thought to the possibility of improving. So having classes or simulators isn't going to help, because they wouldn't take advantage of them.
...unless it becomes a requirement for a license.
The thing is that it's not so much that most people don't know how they could drive more defensively/safer/etc., it's that they're in a hurry, or they're texting, or otherwise aren't carefully monitoring their surroundings.
Yes, there are some people driving who probably shouldn't be on the road. But until such times as automation arrives, you'll find exactly zero support for that idea outside the young and fit of Silicon Valley.
I'd also add that fatalities are far more often caused by reckless young drivers than by the not so fit and young whose reflexes or eyesight may not be what they once were.
Basic driver's education in the state I grew up in was state subsidized and offered through the community college so it was quite inexpensive. On the flipside, it was a terrible course, with minimal actual driving time and a terrible case of design for the lowest common denominator. I wonder if more advanced driving courses could be offered the same way, but perhaps they'd fall to the same problems.
You might look at motorcycling as a positive example. While many states require the Motorcycle Safety Foundation basic course for a motorcycle endorsement, MSF also publishes an advanced course and various skill courses that seem to be relatively popular amongst motorcyclists. MSF courses are developed by the Foundation and all the materials are licensed (I would assume at nominal cost) to various local organizations that actually teach them, in my area affiliated also with the community college.
Therefore, if we're going to require advanced multi-day driving schools and/or annual physical driving exams (or whatever), you're going to have to explain how that works in a world where many people can barely afford to keep their clunker running and are unemployed if they can't drive.
Driving may not be a "right" (whatever that means exactly as we place limits on lots of rights), but I wouldn't want to be the politician who decides to take that "privilege" away from 20% of the population because they missed their annual refresher course or recertification exam.
I think a lot of people would benefit from being able to practice their parking in VR rather than with a >£25/h driving instructor.
https://www.iamroadsmart.com/courses
I am not qualified to have an opinion on whether a pilot can do simulators only but intuitively, an hour of simulator seems to be worth more than one hour of real life flight.
The only difference I can see between a modern simulator and the real thing is lack of consequence.
However, you can't get away from actually flying a plane. There's little things you don't practice in the simulator, like making sure you have your handheld VHF radio, checking for water in the fuel tanks, and how to balance cargo, that you'll never learn in front of a computer screen.
If it's anything like network administration, I'd say something like 20% real experience vs 80% simulated might be a good mix. People who favor just one or the other lack performance under adverse circumstances compared to admins who do both, I've noticed.
From TFA:
> Many of Transport Canada’s aviation inspectors, some of whom are licensed pilots, are not able to keep their licences current because Conservative spending cuts imposed in 2011 curtailed their ability to fly government aircraft. He believes the department changed the rules to ensure federal aviation inspectors maintain the proper credentials.
> A CFPA survey of licensed aviation inspectors, conducted in April, found that about half have not been assigned flight training for more than a year and one in 10 has not been assigned to flight training for up to 10 years or more.
> “With the stroke of a pen, Transport Canada has devised a work-around so their own pilots comply with Canadian Aviation Regulations,” Mr. McConnell said. “Aviation inspectors who are pilots are becoming like traffic cops without a driver’s licence. This situation undermines public safety and inspectors’ credibility with the industry they are supposed to oversee.”
I am sure it has been done based on a serious look at what is important for our pilots to do.
Yeah, except fly a real plane to maintain that pilot's license.
> United States and European Union, which still compel pilots to meet stringent standards to keep their licences valid, including a requirement to actually fly a plane.
Not true, at least not in the US. According to 14 CFR 61.56 (i) "A flight simulator or flight training device may be used to meet the flight review requirements of this section"
> There is a whole host of skills you don’t exercise in a simulator
True, but the opposite is also true. I can think of number of scenarios that can be practiced in a sim but cannot be safely or legally practiced in aircraft. Example of such scenario: engine failure in instrument meteorological conditions.
> If you had a heart issue, would you go and see a heart surgeon that hadn’t operated in five years
I wonder if Mr. McConnell is aware that every single line pilot who files B787 or A380 today did all their training for that airframe and passed the rating test in a flight simulator.
The reality is all airline pilots train in flight sims these day. There is no practical way of doing that in actual aircraft. If one can do all their training and pass the license test in a sim why can't they do periodic flight review in a sim?