Planes without pilots - NYT

The amount of assumptions I hear from pilots about what is essentially drone technology is astounding. Will there be pilotless airliners? YES. Will it happen within our careers/lifetimes? YES. Will it be publicly accepted? YES. Will it be safe? YEA.

My dad was a DoD/DARPA project leader for half of his career with the government. Among other things, he got to design/develop/test drone technology for the USN. The technology is so much further than most people realize. I think at this point it is a matter of infrastructure over technology. The national airspace system needs to be changed before pilotless airliners become a reality. But make no mistake, it is coming.
 
esa17 said:
Single pilot cockpits with ground based datalink control will happen before I retire.
Click to expand...

Incremental steps towards elimination of a huge expense. This will be the first step.

Jafari RastaMan said:
I don't see it happening at the airlines in my lifetime. But if Amazon gets some kind of drone delivery service, I could see it paving the way for cargo carriers (UPS, FedEX, etc.) within 20 years.
Click to expand...

Cargo carriers over low population density routes will be the proving field.

ppragman said:
It's a lot closer than I think most of you really believe. Same with driverless cars - that's 4 to 5 years or less from widespread implementation, and in 20 years good luck insuring a car without it.
Click to expand...

One just completed a coast to coast trip.

http://money.cnn.com/2015/04/03/autos/delphi-driverless-car-cross-country-trip/

RDoug said:
It took a human pilot to make the decision that an Airbus wasn't going to make Teterboro, and that the Hudson River was the only real option. A computer would never have come to that conclusion because it would have been programmed to only accept a runway as a suitable landing site.

I'll stick with a human pilot making the decision only a human would make.
Click to expand...

While true that will be easy to discount as a one off occurrence. They don't need to make a computer flown airplane crashless, they just need a computer that crashes less than a human.
 
The amount of derp in that article makes it almost not even worthy of actual discussion. Here's a wonderful little faulty leap-of-logic, which implies that autoflight and FMSs are somehow autonomous, and that if the pilot's hands are not on the yoke, the airplane is droning on without any pilot input:

Commercial aviation is already heavily automated. Modern aircraft are generally flown by a computer autopilot that tracks its position using motion sensors and dead reckoning, corrected as necessary by GPS. Software systems are also used to land commercial aircraft.

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.

And commercial planes are becoming smarter all the time.
Click to expand...

With a lack of understanding of how the process works that is this staggering, it is just simply not possible to even progress in the discussion of the merits of the idea.

So, let the ignorant opine in the newspaper, but don't plan on it leading to actual policy.

No, the "day of the drone" is not arriving for passenger service in any of our career-spans. Yet again, one need only look at the safety statistics for the world's #1 remotely piloted aircraft operator (in both number of operators and vehicles, as well as hours flown), who has been flying the most advanced types of RPA technology for two decades -- to see that there's no magic safety talisman there without a significant -- significant -- leap in technology and reliability of both the hardware and the datalink infrastructure that it relies on.

Where do people come up with these ideas? The answer to a single madman in an airplane is to create a situation where now there are TWO people (one in the aircraft, one on the ground) and they are both alone? And one of them is going to be able to control the aircraft via a "secure" (yeah, right) datalink?
 
The privatization of space travel will have a lot to do with it yet that's an angle no one has ever thought about. Obviously, datalinks for airliners will have to be beyond line of sight but right now it's about $75k per month for limited access to an existing satellite which has limited bandwidth. We are talking one or two aircraft.

Once SATCOM becomes economically feasible that's ball game for the FO. When the cost of that service is less than the salary of the FO then all of the sudden there will be a new "safety" initiative to eliminate those dreaded pilot errors. The FAA will fall all over themselves to "enhance safety" when the time comes. Flight planning and routing will become much more centralized and greatly reducing the number of close calls in the air. Bureaucratic agencies love centralized setups.

The aircraft of today are already built to accept the hardware required, all it will be is one more magic box in the auto-pilot system.
 
It will happen when AI surpasses human intelligence. Some estimates are as early as 2020. People currently walking around will keep the manned aircraft fellas in business simply because of fear. After a few decades, the only reason manned aircraft will exist is for nostalgia, much in the same way you can pay to take a ride on a WWII era airplane.

Pilots have no need to fret, the demand for pilots is going to be insane and this will be real slow. However, your easily replaceable guys, freight, pipeline patrol, aerial mapping, practically gone overnight once they get the AI up to speed and figure out the NAS rules. The only thing stopping this from happening now is the difficulty they are having with sense-and-avoid.

TL;DR Unmanned aircraft WILL be better than any human pilot and they WILL take our place....a few decades from now.
 
Last edited:
Hacker15e said:
No, the "day of the drone" is not arriving for passenger service in any of our career-spans. Yet again, one need only look at the safety statistics for the world's #1 remotely piloted aircraft operator (in both number of operators and vehicles, as well as hours flown), who has been flying the most advanced types of RPA technology for two decades -- to see that there's no magic safety talisman there without a significant -- significant -- leap in technology and reliability of both the hardware and the datalink infrastructure that it relies on.

Where do people come up with these ideas? The answer to a single madman in an airplane is to create a situation where now there are TWO people (one in the aircraft, one on the ground) and they are both alone? And one of them is going to be able to control the aircraft via a "secure" (yeah, right) datalink?
Click to expand...

The reality has nothing to do with a singe madman. There are billions to be made by the avionics manufactures, and billions to be saved by the avionics purchasers.

The significant technology leaps that need to be made will be made, there are people waking up this morning and getting ready to go to work at Honeywell, Collins, etc. who spend five days a week trying to make it happen.
 
Hacker15e said:
I'm still waiting for my Jetson's-style car that, as a kid, was promised to be by the year 2000.
Click to expand...

haha yeah I think 2020 is a tad early of an estimate. Then you have the legal implications that might slow down the use of superior to human AI for a few years.

What bugs me are the people who say that passengers will never get on an unmanned aircraft. Yes. Yes they will. They already get on unmanned elevators, unmanned trains, and they are currently testing self driving cars with very positive results. As usual, all the technology that short sighted people deny will happen once the computing power arrives.

"You need to learn this stuff because you won't be walking around with a calculator in your pocket!" - every math teacher ever in the 90s.
 
jtrain609 said:
Second: the autopilot on my plane, which is actually really good on most days, commonly can't capture an ILS without trying turn us into an aerobatic aircraft.
.
Click to expand...
The NYTimes can count to potato when it comes to airplanes so that's not even worth addressing.

The problem with your second critique is it is a glib response from an incomplete thought.

Both unmanned aircraft I'm typed in use fully autonomous landing and recovery systems. There is an envelope programmed into the launch and recovery parameters that is very, very predictable. That predictability is the secret. These aircraft won't be vectoring around making decisions the same way a pilot does. They will be on very specific predefined routes that will put the aircraft right where it needs to be for a seamless intercept at the final approach point. How does your auto-pilot do when you're on loc, at altitude, and on speed? Mine does just fine and I'm flying a 40yr old lear.

If weather rolls in then there will be backup routes with the ability to change them on the fly. It takes seconds to do, is exceedingly simple, and is already happening. Traffic flow of unmanned aircraft is actually quite simple when your operators and ATC are on the same page. Once that system is up and running it runs like a top. Imagine what your dispatchers would be like if they could plan arrivals down to the second! It can happen with unmanned aircraft and it's even possible to make the planes keep a set distance from each other virtually eliminating close calls. Unmanned aircraft have near misses for the same reason manned aircraft do, someone didn't follow an instruction. My two close calls were because a manned pilot was in the wrong sector and not because skynet had a glitch.

The infrastructure is easily programmed into the aircraft so integration into the NAS is a non-issue. Keep in mind, most other western nations already do it. Hell, WE already do it. Pred, reaper, global hawk and even target drones have been flying in the US for decades without issue. I can't think of a single manned plane brought down by an unmanned aircraft of any type.

"What happens if there is an engine failure, datalink interruption, the opposite ailerons fail, radar fails, both batteries die and there is a triple gen fail leading to a dual inverter fail!? What then Mr. Big Brother!?" In that situation everyone dies, even if there were three super-captains on board. But in a reasonable set of system failures that require the in depth thinking only a human is capable of? Well, that's why you've got one guy in the cockpit.

Comparing the reliability of Silverfox, Shadow, ScanEagle or even Pred to a 737 is apples to cow dung. They're built with RC parts available at radio shack. Reaper doesn't just fall out of the sky, global hawk doesn't just fall out of the sky. They are much more advanced systems that have much higher reliabilities and far more similar to a Pilatus. Make that comparison and you'll be surprised at their reliability. The one reaper I have seen go down was due to...wait for it...PILOT ERROR!

If there were really a push for it it could happen in under 10 years.
 
esa17 said:
Hell, WE already do it. Pred, reaper, global hawk and even target drones have been flying in the US for decades without issue. I can't think of a single manned plane brought down by an unmanned aircraft of any type..
Click to expand...

Don't shoot your argument in the foot by saying this kind of tripe. It is a massive correlation/causation logic failure to reason that the lack of midairs between CONUS mil RPA operations means that it is all perfectly safe.

There is literally a TFR stood up around Beale AFB every time the Global Hawk is flying. Most Pred/Reaper CONUS ops take place out of Creech (where the RPAs are physically in the NTTR practically from the moment they are airborne), Cannon (lots of high-density traffic out there over eastern New Mexico, right?), and out in the barrens of Grand Forks.

The fact is, most mil RPA ops don't even come close to mingling with civilian air traffic at all by nature of where they're operated.
 
Trip7 said:
Autopilot and VNAV can't even make altitude restrictions on speed yet folks think pilotless planes will happen. Ha!
Click to expand...

"What's it doing?"

*AUTOPILOT*

"It's not doing what I told it to do, that's what."

Also had to TCS the plane the other day when it couldn't hold speed within 15 knots in the climb in FLCH and was going to overspeed.

Pilotless airplanes my ass.
 
You must log in or register to reply here.
Back
Top