Slashdot Mirror
Gaugeless, Computerized Cockpits
CriX writes "There's been some work on creating a more intuitive airplane cockpit. The idea is to use graphics and symbols instead of gauges and numbers to enhance a pilot's awareness. The data shows that these are a lot easier to read under turbulence and the information easier to interpret for less experience pilots. Sweet, I love human factors engineering."
Another thing to break. Honestly, my computer breaks a lot more then my speedometer does on my car. While this is really neat and it probably set out all the things it is required to do, I'd rehter have a simple, 2-3 part blivit telling me info rather then a highly complex LCD.
Objects in the blog are closer then they ap
who pays for all the pilots to be re-trained?
who pays for all the new cockpit instruments?
thats right... I do! (think airline seat prices)
airline industry is not one that can afford to do something like this. would only be practical for boeing or airbus to do on the new planes coming off of the assembly line.
The arrogance...geez. So Colburn suspects that the Kennedy accident was due to the pilot fixating on a single gauge. He thinks that if there was a new display there's no way that there would have been a problem.
Good grief. This is just a bunch of guesswork without any evidence or even any sound reasoning whatsoever. This guy can't make a good pitch for his new technology so he'll claim that some famous accident wouldn't have happened if they had used his technology. It really sickens me when people try to use public tragedies to make money.
GMD
watch this
"The system is designed to preserve the status quo as opposed to bringing forward innovations," Temme said. "We're fighting inertia."
Well, the system is designed to maintain the status quo because it works pretty darn well. All the statistics indicate that flying is the safest form of transportation. Perhaps if (a) Temme was able to show that cockpit confusion was a problem serious enough to warrant a solution and (b) they showed with scientific studies that their new-fangled system improved the situation by a significant amount, perhaps people would listen to them. But just to claim that the reason everyone doesn't drop a system that has worked well for decades and adopt a radically new instrumentation panel is due to bureaucracy is pretty presumptuous of them.
GMD
watch this
easier to interpret for less experience pilots.
:).
This is a perfect example of eliminating the need to understand how something works in order to operate it. A pilot that understands how and why his/her plane works will be a better pilot because they will be able to interpret things about their flight that are not told by the gagues. Are we really better of if we make planes that you can fly without a great deal of knowledge, understanding, and experience?
It's like having an idiot lite for your car's oil pressure instead of an oil pressure gague. The light only gives you 30-40 seconds advanced warning that you're about to destroy your engine, while the gague can potentially warn you of an upcoming problem much sooner (even if the problem is probably a broken oil pressure gague
I wonder how much this has to do with the warned pilot shortage, as many retire over the next few years.
Probably everything.
Joe
http://www.joegrossberg.com
A few comments as a pilot:
This article is SO light on details and substance that I am amazed that it was posted. Instrument flight is difficult, yes, HOWEVER, current guages provide the experiences instrument pilot with a lot of useful information.
A major concern of mine would be that the system designers replace the feedback that the old instruments provide with something that is easier to interpret, but is missing some information content that an experienced pilot would get from the traditional AI,ASI,HI,VSI + Radio Nav / GPS of today.
Flight directors and EFIS displays are excellent today, and, the new large artificial horizon display that can been seen in the Cirrus Designs SR22 upgrade avionics, along with the traditional instrument layout (even if electronic) is a major boon to safety and reliablilty. This also has the advantage of positive transfer of training. Something that the article's system might not have, but who can tell? There wasn't a ton of information in that fluffy article.
People interested in this topic would do well to search for info on the FAA Alaska projects and 'highway in sky' instrument display systems that have been prototyped over the last few years.
As for people who are concerned about failure. Rest assured that even today, aircraft of all sizes that are certified for instrument flight have redundant gauges and systems. Even a B777 has a simple mechanical aritificial horizon and wet-compass hidden in with the electronic instruments.
Hee hee. So let me get this straight, instead of reading three or four different guages that would tell you you're in trouble, now your instrument panel will project an illustration of a plane crashing in flames?
I mean, I can't wait until I can get into a plane and press 'GO' to take off and fly, and then 'STOP' to land.
God bless the military.
--"The perfect example of the man of action is the suicide." - William Carlos Williams
Fly-by-wire aircraft use computers all the time. There are backups for many real-time systems used for this purpose, and after nearly thirty years, these are proven products. The space-shuttle is a very good example of this. Early space-shuttle control systems were a nightmare of guages and displays. Now, it's a very refined system, with an emphasis on what the crew needs to know instead of some instrumentation jockey's wet-dream.
Many new aircraft are equipped with the MFD's (multi-function displays) that the military has had for decades (in one form or another) and it's about time. Most of the work done in this field is done for military use.
Anyone can bitch all they want, but when Betty starts squawking you listen and when you look at your panel you need to know exactly what's wrong. MFD's do that better than any analog guage. Combine advances in data delivery with multiple bioinformatics and a pilot will have a nearly intuitive understanding of what's going on.
I don't see a downside here, since everything is going to be exhaustively tested before it's permitted to move mass quantities of crunchy humans from one shopping spot to another. It's going to be some time before the really cool stuff finds its way into public transportation. Personally, I'm still waiting for a civilian version of a HUD, because what passes for useful car instrumentation blows (looking through the gaps in the steering wheel for data sucks) and from a consumer point-of-view, nothing addresses this shortcomming (not even that bs Cadillac thermal display).
Cheers.
Every new form of media has it's own Requirimento
I am a glider pilot. I must digress. I prefer numbers. Since I have enough hours in an SGS 2-33, I know what the guages are measuring without seeing the numbers. I only need needle position. BUT, the numbers are there when I need them. Seeing the needle at about 7000 feet is good, but I NEED to see 1,500 feet exactly when I start my pattern, or 3,500 to enter the start gate.
As for glass cockpits/ fly by wire, these are no different from normal analog cockpits, except the info happens to be on a screen. The guages, buttons, and controls are all in the same place, and they all look the same. They just happen to be on a screen. Next time you fly Southwest, you will probably fly on both the 737-700(glass) and the -300 (analog). They are the same. The instruments in use now ARE symbolic. The altimeter forms one shape for about 4000 feet, and another for about 14000. But the important numbers are there when I need them. I prefer what I have now.
Sig (appended to the end of comments you post, 120 chars)
Isn't this going to complicate communications with air traffic control?
From the paper:
OZ was specifically designed to reduce, if not eliminate, the need to scan separate and fundamentally different flight instruments. [...] OZ presents information in a graphical fashion. As such, information is processed by the human visual system at speeds it uses to process images, speeds faster than those required to foveate and read dials and gauges and integrate numerical data. We are currently designing studies to measure the rate at which OZ transfers information to a pilot.
[...]we have recently found that trained pilots can simultaneously fly two simulators (each simulator with its own OZ display) executing different maneuvers, under severe turbulence conditions. This demonstration strongly suggests that OZ can be considered to be a single instrument that integrates all the information needed to fly the aircraft.
I can't find the article, because the link now seems to be broken, but Donald Norman has/had quite a lot to say about the interface of aircraft cockpits in his book "Things that make us smart".
From memory though, he was mostly impressed with the current evolution of aircraft control panels rather than scathing of them in any serious way. Among other things, the large and mechanical controls have an important side-effect of aiding communication between the pilot and co-pilot and what each of them is doing at any given time.
His main negative comment, which I think I agree with in principle although I'm not a pilot, was that some of the dials had several needles, and they weren't automatically intuitive to read. Looking at them had to be learned, and it wasn't so much of a surprise that many accidents in the past have been caused by things like pilots mis-reading the altitude indicator.
Much of this could be fixed (in theory at least) by using digital numeric displays instead of dials, but there are other problems. Also base 10 numbers made from arabic numerals arguably aren't intuitive. Interpreting ordinary numbers is another thing to be learned, and although virtually everyone in modern society can read numbers, there have still been measured differences between looking at a number and interpreting what it stands for.
Getting a glimpse of a four digit number could mean something comparably big (9111) or comparably small (2111). Compare this with a graphical representation of the numbers using bar comparisons for example, and there's instantaneous recognition of the meaning without all the brain processing that needs to go on.
Perhaps this is what the intent of the article is