5 ms·

What's the biggest practical limitation to switching to unmanned drone fighters?

saraid216 12y ago
I don't know, but my guess is hackability. I think pilots still have a ton of direct information that wouldn't be significantly vulnerable to hacking, and a lot of direct control that wouldn't be available to drone operators.
zipwitch 12y ago
The Air Force. The biggest limitations aren't practical, they're political.
- TheCowboy 12y ago
  Going into more detail on this would have been more useful, but it is part of the problem. I believe it has to do with the culture of promotion that favors pilots, not to mention the revolving door. The other major part of the problem is that drones are cheaper and present real competition to juicy manned contracts, so defense contractors don't like them. Drones might be an example of actual disruptive technology: a cheap alternative that also creates a new market (planes that can stay in the air longer than a pilot ever could). Losing one drone, or even a dozen drones, is always smaller than losing the life of a single pilot or their plane.
  - tsotha 12y ago
    Drones are far more expensive than people realize. The reason the Air Force and CIA have held off on retiring the U-2 is that it's cheaper to operate than its unmanned competition. Sure, small drones with a video camera and LOS control are cheap, but once you start talking about drones with nontrivial amounts of ordinance and satellite quality sensors you're up into the tens of millions of dollars. For the most part they are cheaper than manned alternatives, but by more like a factor of two than a factor of ten.
    - TeMPOraL 12y ago
      I wonder how much of the cost of drones would drop if they went to mass production. For manned aircrafts, the limiting factor is trained pilots.
      - tsotha 12y ago
        You have to be a pilot to fly drones in the USAF. Not really sure it's necessary, but that's the policy unless they made changes since I looked into it.
        TeMPOraL 12y ago
        It's true, but this is a political, not technological limitation. Politics can change, and there's nothing that precludes creation of semi-autonomous drones (i.e. a single pilot operating a whole fleet) or fully autonomous ones.
remarkEon 12y ago
Had a big argument with a coworker about that this week, actually. What's usually brought up is that having a human being physically in the battle space affords you situational awareness and experience that just can't be replicated, as of yet, with a Drone. What happens if the enemy somehow cuts off the link between Drone and control? You can set the null to fly it back to home or wherever if that happens, but, generally, a human pilot can do a lot more than that - right now. I don't disagree that on a technical level Drones would be able to, eventually, out do a pilot in theory. They aren't limited by having to keep a human alive or conscious in-flight. But there are some pretty scary moral implications if we start affording more and more autonomy to Drones in warfare. The question shouldn't be an "all drones" vs "all pilots" argument. It should probably be more of a "how do we intelligently and operationally integrate the capabilities of Drones into our strategic goals" conversation. That just doesn't seem to be happening with the F35.
tomjen3 12y ago
Lag and poor visibility. Drones work on slow ground attack convoys and buildings where they can fly around for a time but they are controlled via satelites so there is a couple of seconds of lag between when the image is recorded and the orders are received by the drone. In a dog fight, which is really what you have to build a fighter plane for, that kind of lag means you lose every time. Add to that that drones are too slow, have no agility and little to no armour and they are going to be completely useless even against obsolete planes - SAA (Assad) show down one a few days ago. The US has stealth drones, but that is only useful till you are discovered. If you are going faster than the speed of sound that is one thing, but these drones are not.
- Diederich 12y ago
  I wonder how possible it would be construct UAVs that were specialized for air to air combat. They could operate in something of a hybrid mode, perhaps with ground control most of the time, but acting in full automatic mode during the dogfighting. There'd be a lot of advantages. I presume such a dogfighting UAV would have visibility in all directions, all the time. They could potentially pull much tighter turns than any aircraft with a human on board, and with real-time point to point communication, could work together in ways that are impossible with human flown fighters. Part of me wonders if a lot of the amazing game AI that has been developed could apply directly to such a project.
- vacri 12y ago
  Dogfights are a thing of the past. Fighters don't joust at close distances anymore. I did a small part of my degree at a defence research place in the mid-90s. I remember someone showing me an experiment on how best to present range information about other planes to a fighter pilot, in terms of physical appearance. What was curious was the names for the ranges: 'safe', 'lethal', and 'very lethal'. 'Very lethal' was still several kilometers, and I asked about the guns and dogfighting - and got 'that doesn't happen anymore'. More significant issues are speed, weather, and payload size. If you need fighter cover now, you don't want to hear "but it's raining". And speed is needed to increase operational range and response time.
- Symmetry 12y ago
rasz_pl 12y ago
Same as switching Shuttle, or any modern passenger jet to computer controlled landing.

I don't know, but my guess is hackability. I think pilots still have a ton of direct information that wouldn't be significantly vulnerable to hacking, and a lot of direct control that wouldn't be available to drone operators.

zipwitch 12y ago

The Air Force. The biggest limitations aren't practical, they're political.

TheCowboy 12y ago
Going into more detail on this would have been more useful, but it is part of the problem. I believe it has to do with the culture of promotion that favors pilots, not to mention the revolving door. The other major part of the problem is that drones are cheaper and present real competition to juicy manned contracts, so defense contractors don't like them. Drones might be an example of actual disruptive technology: a cheap alternative that also creates a new market (planes that can stay in the air longer than a pilot ever could). Losing one drone, or even a dozen drones, is always smaller than losing the life of a single pilot or their plane.
- tsotha 12y ago
  Drones are far more expensive than people realize. The reason the Air Force and CIA have held off on retiring the U-2 is that it's cheaper to operate than its unmanned competition. Sure, small drones with a video camera and LOS control are cheap, but once you start talking about drones with nontrivial amounts of ordinance and satellite quality sensors you're up into the tens of millions of dollars. For the most part they are cheaper than manned alternatives, but by more like a factor of two than a factor of ten.
  - TeMPOraL 12y ago
    I wonder how much of the cost of drones would drop if they went to mass production. For manned aircrafts, the limiting factor is trained pilots.
    - tsotha 12y ago
      You have to be a pilot to fly drones in the USAF. Not really sure it's necessary, but that's the policy unless they made changes since I looked into it.
      - TeMPOraL 12y ago
        It's true, but this is a political, not technological limitation. Politics can change, and there's nothing that precludes creation of semi-autonomous drones (i.e. a single pilot operating a whole fleet) or fully autonomous ones.

remarkEon 12y ago

Had a big argument with a coworker about that this week, actually. What's usually brought up is that having a human being physically in the battle space affords you situational awareness and experience that just can't be replicated, as of yet, with a Drone. What happens if the enemy somehow cuts off the link between Drone and control? You can set the null to fly it back to home or wherever if that happens, but, generally, a human pilot can do a lot more than that - right now. I don't disagree that on a technical level Drones would be able to, eventually, out do a pilot in theory. They aren't limited by having to keep a human alive or conscious in-flight. But there are some pretty scary moral implications if we start affording more and more autonomy to Drones in warfare. The question shouldn't be an "all drones" vs "all pilots" argument. It should probably be more of a "how do we intelligently and operationally integrate the capabilities of Drones into our strategic goals" conversation. That just doesn't seem to be happening with the F35.

tomjen3 12y ago

Lag and poor visibility. Drones work on slow ground attack convoys and buildings where they can fly around for a time but they are controlled via satelites so there is a couple of seconds of lag between when the image is recorded and the orders are received by the drone. In a dog fight, which is really what you have to build a fighter plane for, that kind of lag means you lose every time. Add to that that drones are too slow, have no agility and little to no armour and they are going to be completely useless even against obsolete planes - SAA (Assad) show down one a few days ago. The US has stealth drones, but that is only useful till you are discovered. If you are going faster than the speed of sound that is one thing, but these drones are not.

Diederich 12y ago
I wonder how possible it would be construct UAVs that were specialized for air to air combat. They could operate in something of a hybrid mode, perhaps with ground control most of the time, but acting in full automatic mode during the dogfighting. There'd be a lot of advantages. I presume such a dogfighting UAV would have visibility in all directions, all the time. They could potentially pull much tighter turns than any aircraft with a human on board, and with real-time point to point communication, could work together in ways that are impossible with human flown fighters. Part of me wonders if a lot of the amazing game AI that has been developed could apply directly to such a project.
vacri 12y ago
Dogfights are a thing of the past. Fighters don't joust at close distances anymore. I did a small part of my degree at a defence research place in the mid-90s. I remember someone showing me an experiment on how best to present range information about other planes to a fighter pilot, in terms of physical appearance. What was curious was the names for the ranges: 'safe', 'lethal', and 'very lethal'. 'Very lethal' was still several kilometers, and I asked about the guns and dogfighting - and got 'that doesn't happen anymore'. More significant issues are speed, weather, and payload size. If you need fighter cover now, you don't want to hear "but it's raining". And speed is needed to increase operational range and response time.
Symmetry 12y ago

rasz_pl 12y ago

Same as switching Shuttle, or any modern passenger jet to computer controlled landing.