April 06, 2015
Airline Pilots Manually Fly Airplanes For 3.5-7 Minutes Per Flight
The Germanwings suicidal/homicidal depressed and narcissistic pilot is providing the impetus to speed up development of remote and robotic ways to control an airplane. Already pilots spend very little time controlling an airplane manually. Boeing pilots work twice as hard as Airbus pilots. Does that make the Airbus pilots lazy? Or the Boeing pilots overworked?
In a recent survey of airline pilots, those operating Boeing 777s reported that they spent just seven minutes manually piloting their planes in a typical flight. Pilots operating Airbus planes spent half that time.
NASA is going for remote control by ground operators - which is how UAVs are controlled today. But remote control seems like a transitional technology until computers on the airplane do it all.
I expect complete flight control automation will have the biggest impact on smaller aircraft where the pilot and co-pilot salaries make up a larger fraction of total operating costs. We'll get more direct flights between small airports once pilots are no longer needed.
Randall Parker, 2015 April 06 09:14 PM
It's not just mentally ill pilots--with so little manual piloting time it's no wonder that catastrophes like Air France Flight 447 happen, where the human pilots have managed to forget the most basic mechanics of powered flight.
That being said, I do hope "more direct flights between small airports"gets a boost from robotics. NASA's Small Aircraft Transportation System (SATS) initiative seems to have quietly died sometime in the last decade.
Pilots are "cockpit managers". Indeed, aside from their flying skills, they are taught this. About the only time a plane is under hand-wield is takeoff. Hinging on the company's policy, landing can be either. Selflanding is common, but many, mainly oldtimers, like better to land the aircraft by hand. Otherwise the aircraft is under auto-pilot. For true, a non-working autopilot will down the aircraft.
There will always be a pilot(s) onboard in the foreseeable future. How many folks would board an aircraft that flew without a pilot? I wouldn't. Too many things go wrong that need a human there to fix or make a fast decision. If you screw up a drone, you'v lost a drone. You screw up a passenger flight, folks die.
Well, sure, if you screw up a passenger flight, folks die. But if it is cheap and convenient to fly between two small cities, (ie, the flights are frequent as well as cheap) then the vast majority of people will fly instead of drive, and total deaths due to travel could very well be lower, even a lot lower.
We need some data here. How often do (will) autopilots fail? How often does (could) an on-board pilot salvage the situation?
If the second number is very high, the sensible solution still probably isn't to have a pilot on every flight, just in case. Probably the efficient fail safe mode is to have a drone link to air traffic control in every plane. If the autopilot shuts down/fails/whatever, or the plane starts losing altitude/acting crazy, some monitor in the cockpit gets triggered, and estabishes a link to the nearest air traffic control station, which takes over and lands the plane in drone mode. .
I can't giv you a number on how often an auto-pilot fails but I delt with it often enuff as a flight dispatcher. If it happens in-flight, that was enuff to shunt them to the nearest airport where we had mechanics. But it would only take one self-flying aircraft to crash or remotely flown aircraft to crash owing the AP failure and no one would fly again. I rode a lot in the cockpit, and there is enuff little stuff that goes wrong to warrant the pilots. If it goes bad wrong, then two pilots are needed to manage the cockpit. We'v gone from a crew of three to a crew of two on most aircraft. I don't think you'll see it go lower. It's not unknown for a pilot to hav a heart attack while in flight so there is a built in redundancy with two pilots on board. I wouldn't fly in a pilotless aircraft nor would I insure it if I headed up an insurance agency.
As with driverless cars, the early adopters will be freight transporters.
For the record, I lol'd at LoboSolo's assertion than a AP failure leading to fatalities would cause people to stop flying. Luckily for us, we've never had an instance of a pilot overriding a perfectly-functioning AP system that led to a crash and passenger fatalities.
I will never get on a plane that does not have pilot. They can fly my Amazon deliveries, but not me.
This may be so, but let's not forget the pilot is there for another very important reason: in Air Emergencies. Remember US Airways Flight 1549? Multiple bird strikes caused it to "land" in the river? An auto pilot or a remote pilot most likely would not have been able to react in time and crash could have been fatal to all on board. Keep the pilots in the cockpit.
I've been retired for 10 years, but throughout my career, unless it was mandated, (ie LAS departures) I hand flew the airplane from takeoff to cruise and during descent and landing. Most pilots I flew with did. With swept wing aircraft it is too difficult to comfortably fly straight and level at jet altitudes manually. I have also flown many Cat III landings both auto land and with the head up device, and I always felt more comfortable hand flown.
I have one word for those in favor of pilot-loess planes: "Sully"
"Probably the efficient fail safe mode"
There is no safe failure mode for cockpit automation. That's why we still have pilots.
Not just failures, can these systems be hacked to replicate 9/11? Are you ABSOLUTELY sure?
Computer reliability: reminds me of how long it took for jet engines to be seen as so reliable that finally only 2 were required for long over-ocean trips. Computers can be made more reliable as well for long term operation.
Jet engine reliability and computer reliability and anticipatory reactions are worlds apart. And even though there are seems of data to the reliability of modern engines, NO commercial single engine aircraft is allowed over water farther than it can glide back to land.