A Machine in the Co-Pilot’s Seat

How long before the robot moves over to the left seat?
I can see two robots flying the airplane (just for redundancy) and a computer operator in the jump seat.
 
I'm thinking it won't be long before military and commercial aircraft are piloted by a guy sitting in a cockpit mockup on the ground. Unobstructed VMC 360° view, better support, better weather reports. It's coming.
 
Well look at the picture in the article and you see the ignition keys hanging on the AI's cage knob.

So how's the robot gonna deal with that?
 
Using robots or remotely piloted aircraft are a perfect fit for the military or possibly the package/freight industry, but I think it will be a hard sell to get your average passenger on an airplane with nobody upfront other than a machine. Of course, with all of the automation being incorporated into some of the newer jets, you could say that this evolution has already happened. Pilots have been reduced to automation managers with an occasional "event" to keep them awake and justify their job.
 
It just sounds like an advanced autopilot to me.
 
Sac Arrow said:
It just sounds like an advanced autopilot to me.
Click to expand...

And it ain't gonna happen because of meat bag redundancy.

You'll get better automated assist, but anyone who thinks automation completely replacing a human is going to happen in a safety critical system has trouble distinguishing Star Trek from reality.

Single pilot airliners aren't going to happen, even if the right seat does nothing but wait for the left to have a heart attack.
 
Sac Arrow said:
It just sounds like an advanced autopilot to me.
Click to expand...

You mean like this one? :D

Airplane.jpeg
 
I think we can agree that the Airbus automation philosophy is effective, state of the art.

Who among us believes that QF32 would have successfully landed with a single pilot on board?
 
I'd be ok with a single pilot supervising a computer that does the actual flying.
 
I think it could crack the "Sit Sleeper" market wide open

sleepers.jpg
 
As long as each robot pilot has to spend 1500 hours in the pattern in a 172 with students before getting an airline job, I don't see an issue.
 
I don't get it... What's the difference between his "red button" and engaging the auto pilot? (other than where the actuators are located)
 
MAKG1 said:
And it ain't gonna happen because of meat bag redundancy.

You'll get better automated assist, but anyone who thinks automation completely replacing a human is going to happen in a safety critical system has trouble distinguishing Star Trek from reality.
Click to expand...

well, given how often critical Enterprise systems would go tango uniform...

:)
 
Will the general public accept a 100 ton aircraft flying only boxes to fly freely over their town with no human in the cockpit?
 
FlySince9 said:
I don't get it... What's the difference between his "red button" and engaging the auto pilot? (other than where the actuators are located)
Click to expand...

I also don't understand what's with all the connections to the actual stick and throttle. Why would you go through all the effort of doing it that way, when normal autopilot actuators are already around and can do the same thing? (Auto-throttles might not be a normal thing, but certainly a more elegant solution exists.)
 
I don't think normal auto pilots are spooky enough. That thing looks pretty darn creepy. :eek:
 
write-stuff said:
I'm thinking it won't be long before military and commercial aircraft are piloted by a guy sitting in a cockpit mockup on the ground. Unobstructed VMC 360° view, better support, better weather reports. It's coming.
Click to expand...
It's been here for a while, more or less. QF-4 through QF-16, X-36, X-48, and many others.

Nauga,
and Ensign Nolo
 
Are pilots that expensive? I thought the problem was that pilots were too cheap. Obviously the guys that have been at the majors a while are making good money but I wonder what an airline has to pay them vs. what it has to pay for the fuel burn of a flight. I'm guessing the cost of the pilot is peanuts in comparison. Actually they took the peanuts away, now you have to pay for those.
 
RussR said:
I also don't understand what's with all the connections to the actual stick and throttle. Why would you go through all the effort of doing it that way, when normal autopilot actuators are already around and can do the same thing? (Auto-throttles might not be a normal thing, but certainly a more elegant solution exists.)
Click to expand...
Because you can have one vaguely pilot-shaped model that fits in many different cockpits. Also, if you read the DARPA solicitation that this was responding to, it's for an 'optionally piloted' airplane - you can strap in R2 or you can strap yourself in the same seat with no interface change.

I can't speak for Aurora's intent (they built the ALIAS model in the article) but pilot controls generally have more control over commercial-level autopilots and stability augmentation - you can do more through the stick, pedals, and throttle than you can with autopilot signals without major model-by-model changes *and* you don't have to buy the autopilot vendor's source code to do it.

Nauga,
sleeping on COTS
 
Given the FAA's conservative, slow moving nature...I can't see something like this happening for decades, if ever.
 
Bonchie said:
Given the FAA's conservative, slow moving nature...I can't see something like this happening for decades, if ever.
Click to expand...
To the best of my knowledge the FAA has no part in the DARPA program the article describes.

Nauga,
and his field expediency
 
nauga said:
To the best of my knowledge the FAA has no part in the DARPA program the article describes.

Nauga,
and his field expediency
Click to expand...

I would think that the FAA has a rep involved. The FAA is already heavily invested in autonomy technology, at least philosophically, with NextGen. This is all part of developing the same capabilities.
 
You must log in or register to reply here.
Back
Top