It looks as though Garmin is dragging the field of aviation very slowly into the 20th century. This caught my eye in a review of the latest version of the Piper Malibu/Mirage: Automatic Descent Mode. From the article:
the M350 also has a Hypoxia Recognition System, active whenever the autopilot is engaged and the cabin altitude climbs above 14,900 feet, as would happen in the event of depressurization. (Cockpit oxygen masks are stowed beneath the copilot’s seat.) If no pilot interactions are detected in these conditions, the system engages Automatic Descent Mode, bringing the aircraft to an altitude allowing recovery from hypoxia.
The system also exists on the Cirrus SR20/22 aircraft equipped with the latest Garmin avionics. (Those airplanes are not pressurized and the pilot brave enough to take them up high is relying on supplemental oxygen delivered to the nose.)
Compared to Siri/Cortana/Tesla this is pretty weak stuff, but it is nice to know that a 20-line Python script is now something that can be approved to protect light aircraft occupants.
[Separately, to see what one driven programmer can accomplish, free of FAA regulation, look at this video about Xavion. The runway length adequacy analysis function could have saved 49 lives and a $28 million aircraft (Comair 5191). The “fly me down to a runway after an engine failure” feature would be a great safety feature if it could be built into a certified panel.]
“a 20-line Python script is now something that can be approved to protect light aircraft occupants”
Scariest thing I have ever read on your blog.
20 lines ? Of the python? What if it descends you into thunderstorm or mountain ? Or another aircraft flying down below? I can see you not used to think about robust Software that handles corner cases.
Another couple of lines of code to make sure you are not descending into a mountain. As for avoiding thunderstorms and other aircraft, if you are unconscious due to hypoxia, you are already at grave risk so a small added risk of flying into a storm or another plane is minor and acceptable. Hopefully the pilot will wake up as soon as the aircraft descends and then he can fly the plane however he wants. It is not always necessary to deal with every corner case if the situation that gives rise to the corner is ALREADY a corner case. Sometimes an imperfect solution is better than NO solution.
People (including pilots) have an irrational fear of accidents that result from outside causes (depressurization) when the vast majority of crashes are due to pilot error. People also consistently overrate their own skills – by self assessment, EVERYONE is an above average driver. There have been a few well publicized crashes due to hypoxia but overall it is an extremely small risk. So if you can do something about it for a nominal sum it is worth it, but mostly it is pandering to the irrational.
Similarly, there is a sizable little business making hammers and other glass breaking devices in case you are trapped in a car under water. What are the actual odds this will happen to you (and that you will be conscious after having driven off a bridge, etc.? But people are very afraid of it. Likewise, all automobiles now have glow in the dark inside trunk releases – how many people each year are trapped in their trunks?
Regarding Xavion, why does the feature have to be built into a certified panel (which as we know will multiply its cost by orders of magnitude when you can just carry an ipad with Xavion on it? Get some velcro or a tablet holder and you now have an additional panel and no FAA to deal with.
BTW, I thought it was funny that when he was showing the “just fly through the hoops” feature, the hoops put him in the clouds so he had to fly outside of the hoops in order to maintain compliance with VFR , speaking of lack of robustness in corner cases.
The fly me to a runway feature seems to again have limited application – most of the time, either you’ll be able to figure out a return path on your own or else there won’t be one available at all.
@Alexey: that’s the beauty of using an interpreted language like Python, as you are flying into a thunderstorm you can just modify the code without re-compiling!
Izzie: He wouldn’t have had to maintain VFR in the event of an actual engine failure. “The fly me to a runway feature seems to again have limited application” — well, let’s hope so!
Why does stuff have to be built into the panel to be sustainably useful? Portable electronics tend to die, e.g., from a lack of charging or battery, just when you need them. You shouldn’t have to go into an aircraft with any more equipment than a pencil and a piece of scratch paper on which to write down the ATIS and clearance.
@Billg
Just make sure you don’t have syntax errors 🙂 would be bad to discover you program that controls airplane stops on the branch that was never tested due to simple typo 🙂
Don’t small planes have a cigarette lighter plugs where you could plug in a charger? You can always carry a USB battery pack Anyway, if Xavion supplements rather than replaces your main panel, then being without it is not tragic – it only puts you back where you started.
Some small planes have lighter outlets. They might be 28V, however, in which case a standard USB charger won’t work. A USB battery pack? Now you’re going to have a loose battery flying around in the cockpit during turbulence? Plus the cables? Have you ever practiced emergency procedures in an aircraft, Izzie? Do you think that when the engine quits it would be a good time to start figuring out how to fly while looking down at an iPad on a kneeboard? How to re-connect the iPad to the USB battery pack? Do you think that if you’re about to take off from the wrong runway that a warning on an iPad on a kneeboard is as effective as one on the primary flight display that is directly under the glareshield? Maybe this would work for you. For me, I would want it in the $200,000 glass panel that I already paid for.
It sounds to me as if your real world choice is to have Xavion built into your main panel X years from now at a cost of XX dollars (where X and XX are large numbers), maybe, if they ever get past the FAA, or else you could have it right now on an iPad for practically nothing.
http://www.mygoflight.com/ sells a variety of arm mounts for the ipad so you don’t have to hold it on your knee. Some of them actually screw into the panel so you don’t have to worry about suction cups coming loose, etc. They also sell 28V usb chargers.
Izzie: Thanks for sharing those links. You haven’t shared your experience actually piloting any aircraft while using these products. If you are right about portable devices having the same practical value as in-the-panel devices then every pilot who spends $100,000+ on a retrofit glass panel is an idiot because the Xavion (or even ForeFlight) app on an iPad on a mount has superior capabilities.
And, yes, regarding the “real world choice”, that was actually my point with the posting. Because that kind of capability is, given the regulatory environment, not practical to include in a certified system, 49 people lost their lives on Comair 5191 and others will lose their lives following future engine failures.
I’m sure that a $100,000 glass panel with Xavion capability (if you could even buy one, which you can’t at the moment) would be better than an iPad on a mount, but would it be $100,000 better? And what about pilots who don’t have an extra $100,000 lying around? Is there room for a solution that provides most of the benefit for a small fraction of the cost?
The FAA philosophy is “only the best” but this philosophy (combined with the liability risk) has basically killed private aviation in this country – it has turned it into a rich man’s game. The number of small planes now (never large to begin with – less than 1 person in a thousand owns a plane) has declined from 1980 while the land vehicle fleet has grown in proportion to the population.