Teaching a computer to refuel an airplane

I was cruising around youtube and saw this video. It pretty much sums up how hard it is for a uav to refuel in flight. the delay is everything with formation flying. So anyway just thought i would pass it on here...

Pretty cool. I like the basket on the arm bouncing around.

a reply to: boomer135

Cant see why refuelling a drone should be a problem considering the advancements made regarding UAVs in the last decade. Obviously it's a more complex operation than i realize. Then again one would think a computer could react and make fine adjustments far faster and more accurately than a human pilot ever could.

I suppose with a tanker and all the fuel this type of operation entails there are still some tasks we consider to be rather to dangerous to trust to technology completely.

well weve been doing it a while now. The problem has always been that small lag time between the uav communicating with its ground based controllers. but with nearly full autonomous uav's coming out, there isnt much of a lag time and it can be done safely.

a reply to: boomer135

It seems to me that adding three to four small jointed winglets to the drogue would help stabilize it. Those could be hooked up to a computer in the refueler which could make real-time readjustments.

Or has that been tried already?

originally posted by: TDawgRex
a reply to: boomer135

It seems to me that adding three to four small jointed winglets to the drogue would help stabilize it. Those could be hooked up to a computer in the refueler which could make real-time readjustments.

Or has that been tried already?

I've never seen it done but here's my take on it. The MPRS wing pod drouges are far more osculating than the centerline boom to drouge adapter. Something to do with the receiver aircraft probably being right behind the tankers engine during refueling I would think. But the early tests with UAVs were done with the centerline drouge. And they had a much easier time programming the UAVs to make the plug. From what I understand though that video is a little bit old and the technology has already been incorporated into some UAVs.

a reply to: boomer135

Hopefully they do better than the F-18 pilots have done over the years.

I was talking to a RAAF guy in Spain a while back who was working on the KC-30A - he was saying the boom on the Airbus was hugely different to current booms, much more automated..I didnt get chance to probe (sorry) any more.

Anyone know what that means...could that reduce the workload on the pilot/UAN computer?

(sorry Boomer, your kind may be going the way of the pilot)

originally posted by: boomer135
I was cruising around youtube and saw this video. It pretty much sums up how hard it is for a uav to refuel in flight. the delay is everything with formation flying. So anyway just thought i would pass it on here...

Nice video, Ive been to that factory.

I wish they did Science like this when I was at school, no wonder we are getting smarter!

Does the pilot judge the delay currently, are the adjustments he mentioned, from flying behind the engine or from the atmosphere really random or pretty uniform? Im thinking like a wave is a wave - it can be bigger or smaller, but if you dropped 100 ball bearings of the same size into a pool you could probably program something to stay in the same place because you could program the delays.

Similar to an aircraft, you have some knowns - aircraft size, altitude, wind direction, speed. So if your fueller is monitoring all these it could send controll signals to the UAV - perhaps you dont even need to programme the UAV - you have the controll sofware on the refueller and it takes over and flies the uav.

You could even have a pilot in the refueller who takes over and flies the UAV....

originally posted by: IamSirDrinksalot

originally posted by: boomer135
I was cruising around youtube and saw this video. It pretty much sums up how hard it is for a uav to refuel in flight. the delay is everything with formation flying. So anyway just thought i would pass it on here...

Nice video, Ive been to that factory.

I wish they did Science like this when I was at school, no wonder we are getting smarter!

Does the pilot judge the delay currently, are the adjustments he mentioned, from flying behind the engine or from the atmosphere really random or pretty uniform? Im thinking like a wave is a wave - it can be bigger or smaller, but if you dropped 100 ball bearings of the same size into a pool you could probably program something to stay in the same place because you could program the delays.

Similar to an aircraft, you have some knowns - aircraft size, altitude, wind direction, speed. So if your fueller is monitoring all these it could send controll signals to the UAV - perhaps you dont even need to programme the UAV - you have the controll sofware on the refueller and it takes over and flies the uav.

You could even have a pilot in the refueller who takes over and flies the UAV....

To your first question, he's talking about a fly by wire boom, similar to the kc-10s boom. But they can't seem to keep it from falling off the jet inflight (different story). The KC-46 boom is a new generation of fly by wire boom, more advanced than the kc-30s.

The KC-135 isn't fly by wire and I like it better because the controls are exactly what you feel when you fly it. And there is no pilot to judge the delay because they are automated. During air refueling it would be like setting a program for the UAV to use its sensors and other crap to autonomously fly to the tanker and refuel.

And by the way, there isn't a delay anymore. That video was a little old. Somehow they programmed them to do it without the delay anyway.