12

If not, why? Doesn't it make flying safer? It seems to me that automating a small plane should be easier than automating a car (excluding dealing with Air Traffic Control). Is it too expensive?

Pondlife
  • 71,714
  • 21
  • 214
  • 410
Retired account
  • 327
  • 3
  • 9
  • 3
    My 1978 Cessna 177-B has an autopilot, an STEC System 50. – Ron Beyer Jun 29 '17 at 04:17
  • 2
    Note that your comparison to automating cars does not take into account that plane automation has been around for over 100 years (the first autoleveler was used in 1912). While car automation had the earliest ideas in the 1920s, most ideas until roughly the 1980s involved remote controlled cars or cars that require some sort of guidance circuit INSIDE the road, effectively requiring the replacement of millions of miles of road. – Nzall Jun 29 '17 at 07:50
  • 2
    It should be mentioned that current automation in civilian aircraft are more "target value following" (e.g. following a predetermined route) than "decision making" (e.g. adjusting route for favorable winds or to avoid turbulence), so if your example for car is a driver-less car, then there is no autonomous aircraft yet, except experimental or military autonomous drones. – mins Jun 29 '17 at 10:10
  • The question of whether or not automation makes an aircraft—much less a small, single pilot aircraft—safer is a very complex and broad question. If it is answerable at all in this context, it should probably be a separate question. – J W Jun 29 '17 at 11:13
  • @Jonathan Walters but also an interesting one. I'll post it as a separate question – Retired account Jun 29 '17 at 14:48
  • Your comment: I was wondering if there are systems which could theoretically fly without Assistance from after takeoff to before landing, doing everything a pilot can do, like avoiding turbulance, Physical obstacles and dealing with emergencies." is very different than what your current question asks for. No current answers deals with this aspect (but they are upvoted). Please edit your question to clarify. – mins Jun 29 '17 at 15:00
  • Your question sounds like you're referring to Tesla's autopilot, or the ability to guide itself without any pilot input through complex situations. Planes already have features roughly analogous to Tesla's autopilot, but fully pilotless planes are still a long ways off.See this question for more details: https://aviation.stackexchange.com/questions/1802/why-do-we-still-use-pilots-to-fly-airplanes – Cody P Jun 29 '17 at 16:21
  • There's some terminology in aviation to be aware of, with an "FMS" navigational computer that let's you enter navigation points and from that computes flight segment paths, a "flight director" that displays what attitude to hit to fly the desired path, an "autopilot" that auto-trims and also computes the way the plane should fly to follow either the flight director commands or the FMS commands, and a fly-by-wire system that moves airplane control surfaces to reach the desired roll rate, vertical speed, etc. There's some overlap in the last three systems. – Cody P Jun 29 '17 at 16:31
  • @Cody Thank you for the reference. That thread essentially answered my question. Out of curiosity, can you give me a comprehensive reference(s) to learn about aircraft design and operation? – Retired account Jun 29 '17 at 16:42

4 Answers4

15

Many small, light aircraft are equipped with an autopilot today. They can range from very simple autopilots, without altitude selector and so on,

ST SYS 30

Source, ST SYS 30, $ 12,000

to advanced systems with altitude selector, automatic trim and so on.

Bendix/King KFC 225

Source, Bendix/King KFC 225, $ 15,000 - 42,000

And with this GPS, feeding the autopilot with data, it has the abilities of an airliner.

Garmin GTN 750

Source, Garmin® GTN™ 750, $ 17,000

You have to know, the systems in the general aviation industry are getting better and better. Of course they are way below the complexity of the avionics of airliners, but the features are the same if not more.

For example: A captain of an airliner has to review or change the flight plan with the FMC, respectively MCDU on Airbus, a not very straight-forward device to control the plane.

Of course, the FMS has a lot more capabilities than a GA GPS, but one of it's primary features is the input of the flight plan. And in that, from the usability-view, the Garmin® GPS has the better solution: An all-in-one navigation device with touchscreen input.

Noah Krasser
  • 8,992
  • 3
  • 50
  • 100
  • 2
    What an incredibly lucrative software field. – Fattie Jun 29 '17 at 16:35
  • 2
    I don't know that it's lucrative. I'd guess it's expensive (high cost to develop and, especially, to QA, and maybe even to sell or install) and perhaps few sales. – ChrisW Jun 29 '17 at 17:49
  • @Fattie If you are interested, I have already asked a question on this: https://aviation.stackexchange.com/questions/38122/why-are-general-aviation-avionics-so-expensive – Noah Krasser Jun 29 '17 at 19:23
  • 1
    great QA @NoahKrasser, thanks. I can tell for a fact that in one of our particular software specialties, the company, which, does "the same as everyone else" but operates in the "aircraft instruments" field - makes spectacularly more money than the rest! :) – Fattie Jun 29 '17 at 19:32
  • @ChrisW Not to mention high risk of liability if you get anything wrong and with software, you almost always do. – David Schwartz Jun 29 '17 at 21:11
  • @Fattie No, "Angry birds" is an incredibly lucrative software field. If you ever get involved with developing safety-critical software, you will find out why these guys don't make much money at all. But if they cover their costs, getting the product out there is very well targeted advertising that raises brand awareness for their commercial-aviation products. – alephzero Jun 29 '17 at 23:37
  • 2
    @Fattie: As with anything in aviation including software: if you want to make a little money start with a lot of money. Getting things approved and certified is expensive and can cost more than all your initial investment + developer time. And the thing is, it is legally mandatory - you cannot sell it if it's not approved. Chinese companies like Yuneec have been trying for years finally giving up and selling small drones instead (almost no manufacturer regulations, all regulations regulate owners) – slebetman Jun 30 '17 at 00:13
  • hi @alephzero, you'd be surprised, behind the scenes one of the biggest earners on the "app store" is a serious aviation product. folks are always amazed to learn that Scrabble (yes - Scrabble!) is one of the biggest earners on that venue also. – Fattie Jun 30 '17 at 00:22
  • @slebetman - right, regulatory is the gateway – Fattie Jun 30 '17 at 00:22
14

Yes. Small airplanes can have autopilots. They are very expensive (~$20,000) but are occasionally added to planes as aftermarket modifications or as updates to existing systems.

Autopilots in small planes range from simple wing levelers to fully coupled systems that can use GPS systems fly instrument procedures all the way to published minimums (200' above the ground in some cases).

I owned a small four seat plane with an autopilot that could hold altitude and a heading -or- a VOR/LOC course -or- a GPS track. It was considered a relatively basic system.

acpilot
  • 6,016
  • 1
  • 16
  • 35
6

Does it make flying safer? That is a very interesting question that is open to debate. The great thing about an autopilot is that when turned on, it frees up the part of your mind you are using to control the aircraft to actually think about other things such as where you are going. The bad thing about autopilots is that they are complex and need to be set correctly (and can fail). So you need to be more familiar (or current in pilot speak) with the make and model of autopilot, GPS, whatever, in your aircraft. Years ago when I learned to fly the CAA here in the UK didn't encourage the use of Autopilots or GPS. Should they? Like I said it cuts both ways. Before I get loads of comments saying what a dinosaur I am etc etc. I would like to refer to one example I read about in a light aircraft where the autopilot was set to altitude hold in cloud and the engine failed. The autopilot held the altitude and the aircraft stalled, spun, and crashed. For sure they have their place but I just don't get the button I can press and then I don't need to do anything mentality. Maybe thats just me.

Philip Johnson
  • 1,815
  • 12
  • 15
  • 1
    The kind of autopilot I was thinking of would detect the failure of the engine and take immediate action. What you(and I think the other answers too) are describing is more like cruise control in cars than the full automation I was imagining. I was wondering if there are systems which could theoretically fly without Assistance from after takeoff to before landing, doing everything a pilot can do, like avoiding turbulance, Physical obstacles and dealing with emergencies. A plane equivalent of what Tesla is trying to do with their cars. – Retired account Jun 29 '17 at 14:45
  • @A.W.Grossbard - how exactly would you propose the Autopilot handle engine loss in a single engine aircraft? Are you going to map the world and pre-score it for off field landing suitability? What if there's a ballgame in progress or a schoolbus making its way down your highest scored automatic choice? – Chris Stratton Jun 30 '17 at 03:43
  • @Chris As you can probably tell, I have no experience with aviation. But my guess is that the autopilot would take whatever actions are necessary to enter into a controlled descent, identify potential landing/ crash landing sights, select one, and direct the aircraft towards it. You would not need to "prescore" the world for this, you could use google maps to find flat and inactive areas(google actually has live use data collected using Android). And I really don't think pilots are capable of detecting whether a ballgame is in progress and adjusting their flight path as their plane crashes. – Retired account Jun 30 '17 at 04:03
  • @A.W.Grossbard the fact that you have no idea what you are talking about is overwhelmingly evident. Try a discovery flight sometime, it's not expensive. Should you take lessons, what to do when the engine stops will be a practical part of training from relatively early on. Far before anyone would propose automating that, they'd probably just install the existing technology of a ballistic parachute for the entire light aircraft. – Chris Stratton Jun 30 '17 at 04:15
  • @Chris They're like $150 dollars(from cursory googling)! That's too expensive for me at the moment. If you wouldn't mind telling me, what part of what I wrote is obviously insane? There is a standardized procedure for dealing with engine failure and crash landings, right? I'm guessing your thinking that the "just use google maps" part of my comment is stupid, because it is. Disregard it. I get that not killing tons of people when you land could be an issue for the computer to deal with.Do you by chance have a general aviation reference you could refer me to to learn more? – Retired account Jun 30 '17 at 04:29
  • @Chris That actually makes flying seem more safe to me. Thank you for taking the time to respond to my question and comments. I appreciate your expertise and just want to learn more. – Retired account Jun 30 '17 at 04:34
  • Yes autopilots are like cruise controls that follow preset paths. There are some exciting developments like having the screen in front of you show you the scenery outside with the runway on it, even though you are in cloud so you don't have to do all the maths of working out where you are. To me that is amazing, but as far as having autonomous aircraft without a pilot, that blows my mind. Even if it could be done in what circumstances will it sacrifice me to save more than one person on the ground? I think technology is a good as a servant, I'm not sure I'd like it in command personally... – Philip Johnson Jun 30 '17 at 15:53
3

To answer the why part of your question, turn it around. Most of us who fly small airplanes do so at least in part because we enjoy doing the flying. Why on Earth would we want an expensive device to keep us from doing what we are paying good money to do? We might as well just fly commercial :-(

Indeed, I honestly don't see why Tesla is wasting all that money trying to develop self-driving cars. It just ensures that I'll never buy one.

jamesqf
  • 2,577
  • 1
  • 16
  • 13
  • 2
    Well it's fortunate that you are capable of driving reliably, but there are some people who have limited vision or physical handicaps. Not to mention the many people killed by drunk drivers each year. It's definitely not a waste of money. – David Schwartz Jun 29 '17 at 21:12
  • 3
    I will add that most people drive cars just to get somewhere. In that respect, an automobile autopilot is not only desirable to the person, but desirable to everyone else from the elimination of a distracted driver. If they get it to work... an automobile autopilot has to deal with a lot more potential obstructions than an aircraft autopilot. – tj1000 Jun 29 '17 at 23:00
  • 1
    "We might as well just fly commercial" - and some "commercial" pilots think exactly the same way. My employer's chief pilot is type-qualified to fly most of the current Boeing and Airbus range, but what he really likes about flying is spending a Sunday tow-launching gliders at a local club, or doing an aerobatics display somewhere in the company's WWII Spitfire, not spending hours monitoring the glass cockpit display in an A330 at 35,000ft. – alephzero Jun 29 '17 at 23:47
  • @tj1000: Sure, the automobile autopilot would be attractive to limited markets, but hardly to the automobile enthusiast who's the typical buyer of upscale cars. (I won't even buy a car with automatic transmission.) Nor do I think it would help much with drunk drivers, as the drunk would have to be sober enough to decide to turn on the autopilot. – jamesqf Jun 30 '17 at 18:36