US aerospace firm Rockwell, prominent in the unmanned aircraft biz, is engaged in developing truly automatic pilots - ones able to conduct even landings without human input - for light manned aeroplanes. The developments offer another small step towards the long-desired flying car. The company announced the developments at the …
Every flight I've been on that's had an auto pilot landing has been the smothest landing i've experienced. People pilots just seem to rock the plane from side to side to side as tho they're strafing in Unreal Tournament.
Does this make anyone else a bit queasy?
Even assuming that it isn't running Windows XP Embedded, the thing that crashes my internet radio periodically, I'd still prefer a real meat pilot (or two) in charge of anything large and heavy passing over my fragile body.
*Obvious comment alert*
I can see Michael O'Leary rubbing his hands with glee. Now that check-in is automated, all that remain are the take-off and landing.
That said, it'd probably be a lot smoother.
Great - I've been on a hands off landing on a 747, but I'm glad I did not know until the pilot told us when we were safely down. It was as smooth as glass.
it's a shame they didn't employ the same technology to stabilise the cameraman during the landing, i was starting to feel a bit queasy there for a minute.
Paris because she knows how to handle her undercarriage!
This certainly is an achievement. Small aircraft are actually quite efficient (I've heard of Cessna 182's getting the equivalent of 15mpg, even with their 50 year old engine tech and poor aerodynamics), and with a cruising speed of approximately twice that of motorways, light aircraft operating safely out of small airports, high speed point to point transport is a real possibility. The big problem with trains is getting from remote areas to other remote areas - Edinburgh to London is very fast on a train (as fast as a light aircraft would be anyway), but going to any of the outlying areas (Shetland, the Hebrides, Pembrokeshire, Cornwall) is considerably slower. Being able to travel to your nearest light aviation airport (most cities have one), climb into an aircraft that will take you to your destination with minimal input, and arrive after less than an hour for each 100 miles travelled, regardless of terrain or infrastructure, is a very appealing prospect.
It will take a while for this to truly be a safe system. There's a need for a computerised aircraft to have sophisticated traffic avoidance systems in order to minimise the chance of a collision, and certification of any kind of complete automatically piloted aircraft is probably 20 years (at least) off, but there's a lot of potential.
Re: Does this make anyone else a bit queasy?
No, not at all. Having had some involvement in post-air-crash management it is (near as damn it) always pilot error that causes crashes. Machines are generally far more reliable for one simple reason. If one goes wrong it makes HUGE headlines.
There are three things preventing machines from taking over most day-to-day driving/flying/sailing tasks;
1) Technological Development
3) Public "unease" and a general (IMHO misconception) that it's far better to have a human up front
Of those, I'd suggest No. 3 is by far the biggest barrier and probably the thing stopping sufficient funding to overcome No. 1.
But I'd far rather have something driving or flying me around that I could guarantee wasn't
Light aircraft don't take kindly to fast 3-wheel landings like that all the time. You're supposed to come in much more slowly and then lift the nose during the flare to land on the main undercarriage and put as little weight as possible on the nose wheel.
Besides, doing the landing by hand is fun.
That's 'cos the autopilot can react to very small changes in the aircraft's attitude with very small corrections to the control surfaces before the human pilot would ever notice anything. The human pilot only reacts once a deviation to the desired attitude is apparent (either on the instruments or by "the seat of the pants") and often overcorrects, hence the waggling effect.
It's a similar effect to using the Cruise Control on your car. The car will waft along with the speedo needle apparently riveted to one point on the dial. You try and do that yourself for any significant length of time.
"There's a need for a computerised aircraft to have sophisticated traffic avoidance systems"
Supposedly they're working on this in the US to allow military RPVs to fly in the national FAA traffic system. I'm sure the technology will be cheap after the military works it out... *cough*. *cough*
Wheren't the Soviets landing Yakovlev 38s on the decks of their kiev class carriers using a hands free computer-controlled telemetry system run from aboard the ship, back in the late 1970s? Admittedly, the Yak 38 was a useless aeroplane in almost every other respect, but it was still quite a good demonstration of autopilot landing in less than ideal conditions, given that it was using 1970s Soviet computer equipment.
Aint that the 'merkan TV series where they burn a hole in a map in the opening credits? So you don't know where the fuc*k it is? Probably Kansas. Too many yellow roads.
(Bet Hoss Cartwright wouldn't fit in that bugger...Fuc*k'd if I'd trust my life to a model aeroplane enthusiast. Especially using CB - Childrens^W Citizens Band - 27 megs.)
Ok, pub's closing. Gorrit.
The last time I went to the Canary Islands (on a 767-200 I think, drank much since then) the captains voice came over the speaker ".... because of bad weather we can't see the runway, but the autopilot can, so we'll let the plane take the strain and land itself".
The only thing that indicated we'd actually touched down was the brakes coming on. Didn't feel the touchdown at all. And that was bad weather! There was a certain level of panic tho in the cabin, so maybe announcing auto-landing wasn't a good thing.
It's reliability that matters. What happens if the system fails? Autopilot+human is two pilots, of differing competence. That's a good thing.
I wouldn't want two identical machines. It's like having two watches--how do you know which one is right? Getting to the point of a reliable system is going to be a lot of work, but we have the example of Fly-By-Wire systems as a guide.
And if we need automated Air Traffic Control, how does the plane cope with that failing? What ground equipment is needed for a safe landing? And how much does a human need to know?
We're still making it harder to pass the driving test. Which comes first, the driverless car or the pilotless plane?
New and improved?
This article makes this sounds as though it's something new being developed for light aircraft. Of course, all commercial aircraft are required to have full autopilot.
Regards automated ATC: this would be easy. It's virtually there now. The only problem would be the classic big companies + government involvement + new computer system = not going to happen.
All that money...
...and they couldn't afford a proper cameraman. It's almost MTV stylee!!
"I need you clothes, your boots, and your motocycle.."
sent me off to sleep nicely. good story, but man, what a boring video.
<<It's like having two watches--how do you know which one is right?>>
Or, three frozen pitot tubes. AF477?
Screw that automatic mularkey*, I'm walking to US next time. When the Bering Strait freezes over, natch.
* I was on a Finnair flight once, Oulu to Helsinki, and we had a rough-as-conkers landing. They'd just replaced the MD82's for Airbus 320's. At the gate the pilot apologised, and said it was his first passenger-laden automatic hands-off landing. (Obviously he'd done it before, simulator, cargo or unladen) Blamed it on the autopilot. No-one hurt, but it certainly wasn't smooth. Bounced all over the shop.
Beer icon, 'cos I needed one after that trip.
Actually, hands-off landings are a relatively recent thing in commercial aviation and rely heavily on specific installations on the runway one is attempting the landing on.
This is something new.