3
$\begingroup$

Have there been any 'proper' field trials of Automated Air Traffic Control?

Specifically, something that has been tested on larger aircraft and has come beyond the drawing board. By automated I'm implying that all decisions on movements are made by computers and instructions would be for instance transmitted digitally to the cockpit.

$\endgroup$
4
  • 1
    $\begingroup$ There's a book, whose name I've forgotten, by Arthur Clarke about the British efforts during WWII to develop safer landing systems. For one, they used radar to track an incoming plane, and the ground controller told the pilot what to do. On an early test it worked fine, until the plane headed off in the wrong direction and didn't respond to new instructions. Turned out they were tracking the wrong plane, which just happened to be doing roughly what they were looking for. The actual plane was out over the ocean, and the pilot was getting nervous about instructions that didn't make any sense. $\endgroup$ Jan 28, 2015 at 16:11
  • 1
    $\begingroup$ I suspect that this will really start to happen as a 'side effect' of integrating drones (UAS) into the air traffic system. The current ATC approach won't work for unmanned systems, and as drones become more numerous, autonomous and fly at higher levels some kind of automated control will be needed. It may still take decades, but it's hard to see it not happening eventually. $\endgroup$
    – Pondlife
    Jan 28, 2015 at 17:20
  • $\begingroup$ even drones are remotely controlled by a human operator for takeoff and landing, whether or not the launch/landing sequences are automated in software. the operator still has to get proper tower clearances. $\endgroup$
    – Erich
    Jan 29, 2015 at 6:31
  • $\begingroup$ @PeteBecker: That specific method is called PAR. $\endgroup$
    – Vikki
    Apr 23, 2019 at 4:28

3 Answers 3

3
$\begingroup$

No

Short answer, but there has never been anything resembling a "proper" trial. There have been a few small scale, limited and tightly controlled trials, but nothing over any significant time frame, involving a significant number of aircraft, or even involving aircraft in "normal" use.

$\endgroup$
3
$\begingroup$

There have been many trials involving large aircraft and various levels of ATC automation. However these trials do never involve more than a handful of aircraft. And always with a human Air Traffic Controller tightly in the loop for safety reasons.

And never were all decisions made by the automation system. Progress comes in small steps.

$\endgroup$
1
  • $\begingroup$ Examples/references? $\endgroup$ Aug 26, 2015 at 22:58
1
$\begingroup$

I suspect one of the bigger issues with automated ATC is you will always need human controllers to be there to take over when the automated system fails.

Moreover, since the job of an ATC controller is a difficult one that requires a great deal of mental dexterity to translate a two dimensional display or moving object information into a mental picture of a three dimensional airspace, the individual really needs to be active regularly enough to keep those skills sharp.

There is no doubt a computer can monitor and extrapolate how traffic is moving relative to one another far better than a human can, but the whole communication with a human thing is another story. It might work great as long as everything was operating routinely, but odd things happen quickly.

As such, you sort of want it automated but at the same time... you don't.

I rather suspect, and hope, that instead of going down some fully-automated route, the tools presented to the human ATC will improve instead.

How that blends into moves to use pilotless aircraft.. I have no idea.

$\endgroup$
1
  • 1
    $\begingroup$ Much the same way most air transport category aircraft can land themselves via auto-pilot, but the pilots don't just sit back and read the paper while the plane does all the work. Eventually, the computers will fail and when that happens the pilots need to be in practice for when they do. Therefore, the pilots fly most landings themselves. $\endgroup$
    – FreeMan
    Oct 17, 2017 at 18:48

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .