Drones will just be jammed if they go up against anyone with reasonable tech. It will require autonomous AI pilots to make drones sensible challengers for air superiority, and even if the tech is there, the laws and cultural attitude shifts aren’t. Not yet, anyway.
I haven’t kept up on how comms are kept with drones, but I’d imagine if it is radio based they are using frequency hopping
I guess that’s a reasonable approach to take. I’d hate to be the F-35 pilot up against a drone you haven’t been able to jam, though.
Getting an AI to handle that level of interaction wouldn’t be easy. And you’d probably just be shooting missiles at each other anyway so it’s not like there would be a huge difference.
The only real advantage it would have would be G-force resistance. An AI would only be restricted by the airframe in that situation, which would be an advantage over a human in a dogfighting situation. I guess reaction time, but being able to react super fast isn’t that huge of an edge when physics comes into play, inertia still exists, though you could probably have some sort of flexible gun mount to allow it to matter more, in theory.
That makes sense. But, at the risk of wading into an area I know nothing about, wouldn’t the main advantage be ‘numbers?’ For the price of an F35, it seems likely we could field a whole bunch of drones.
I think that gap is closing faster than we think though. I figure once jamming starts the drones will go autonomous and kick our ass, SkyNet style.
To have a drone capable of handling an F-35 you’d have a drone that was likely just as expensive, probably more so on the maintenance end. You aren’t going to be using Predators or the like for air superiority - you need to use air superiority fighters or your drones will just be target practice for the enemy. Which means something like an F-22/F-35 air frame.
I don’t think the advantage is just G-force resistance. In the article I linked, the instructor was astounded by the AI’s ability to read his actions.
I can imagine fighting a drone would like fighting an aimbot but worse - perfect accuracy, perfect response, perfect execution, never tired or makes a lapse in judgment and will absolutely not stop. /terminator
Well, he was also dealing with missiles-only engagements in a simulator without realistic aerodynamics. The real world is a completely different animal where a hundred things can go sideways.
I mean they might get there, but these sorts of situations tend to actually be really tough for AIs to manage well.
I think a lot of the cost of the F-35 comes down to wanting to keep the pilot alive. You want to win engagements and not lose the aircraft. With drones you can pursue a more attritional strategy. I think that opens up the possibility of a much smaller, less capable and cheaper aircraft. It doesn’t need to be able to handle an F-35 one-on-one, just needs to be capable of favourable exchange ratios.
I expect it should be fairly easy to jam a frequency-hopping drone given even modest technical expertise. All you have to do is monitor it for a bit and/or see how it reacts when it is jammed, depending on what its countermeasure is. Moreover if you have plenty of broadcast power you can just overwhelm a very wide range of channels all at once. The transmitter for the drone controller will be much further away than the jammer, so it shouldn’t be too hard to crush it with noise.
Apart from not having to worry about keeping a pilot alive, I think a big advantage of drones is low cost and disposability. Putting g-force aside a non-drone cockpit with all its avionics and intelligence support for the pilot and life support and ejection capability and all that is an 8-figure cost difference compared to a drone. And then since you don’t care nearly as much if they get shot down, what the hell, forget stealth and any ECM that noticeably affects the bottom line. Just make it small, fast, and cheap and crank out a zillion of them.
As Shiva pointed out, you can’t make them “cheap” and have them do what they need to do. Just the sensors on the aircraft are gonna cost millions.
You’re not going to be able to lower the cost to the degree you seem to expect, where you can just throw tons of them at the enemy and have the same effect, where losing them won’t matter.
Well first of all I don’t believe drones will be used in place of piloted aircraft for a generation at least; I’m not an advocate to begin. I’m just considering cost and doctrinal attitude assuming that autonomous drones are eventually a standard choice.
So by “cheap” I mean cheap in comparison with an F-35, (which is after all feasible even for piloted machines given the obscene cost of a F-35). Given a deliberate disposable attitude toward the machine I think they can indeed be relatively cheap. You still need high-quality targeting avionics and a powerful engine but you don’t need all those systems I listed meant to support pilots and you can also reduce your quality standards because your goal is explicitly quantity.
So perhaps the vehicle will be $10 or $20 million instead of $100-$300 million. And despite its cost the F-35 is also very low quality in terms of maintenance and operability. At present you have to buy 3 of them to keep 1 in the air, even for the land-based version.
I doubt you could make anything for $20 mil that would be anything other than target practice for the enemy.
The missiles themselves would approach that cost, much less everything else. Reapers, which are glorified remote controlled airplanes cost over $14 million a piece. And those aren’t remotely autonomous, the cost per unit of a completely new type of aircraft would be likely F-35 levels or close to it.
Alrighty, here are some costs for you.
AMRAAM-D: ~$2mil ea. (source)
AN/APG-83 (advanced AESA radar designed for F-16 upgrades): ~$2.5m ea., per Taiwan contract costs.
(source)
The overwhelming majority of the cost in a defense acquisition program goes to development and integration, not cost-of-systems.
I think that those numbers alone kind of support the case that just making “disposable” UAV’s isn’t really feasible, if you’re trying to do what we’re talking about here.
You’re talking 4 million dollars for a radar system and ONE missile.
Here’s a more detailed breakdown of the stuff in the F35:
http://www.bga-aeroweb.com/Defense/F-35-Lightning-II-JSF.html
Price/Unit Cost: The unit cost of the F-35A is $109.88 million (recurring cost) in FY 2016. The airframe costs $64.47 million, the F135-PW-100 engine costs $13.06 million, the avionics cost $16.74 million, whileother costs make up the remaining $15.61 million.
The unit cost of the F-35B is $121.33 million (recurring cost) in FY 2016. The airframe costs $71.81 million, the F135-PW-600 engine (coupled to the Rolls-Royce LiftSystem) costs $30.82 million, the avionics cost $16.33 million, while other costs make up the remaining $2.37 million.
The unit cost of the F-35C is $117.83 million (recurring cost) in FY 2016. The airframe costs $86.09 million, the F135-PW-400 engine costs $13.06 million, the avionics cost $16.36 million, while other costs make up the remaining $2.32 million.
Due to efficiency gains and process improvements, Lockheed Martin has been able to cut the number of labor hours required to produce an F-35, from 153,000 hours a copy in 2011, down to 50,000 hours by mid-2015. The company expects labor hours to drop even further to just 35,000 hours per airframe by 2020. These reductions should make Lockheed Martin able to offer the F-35 at a unit cost around $80 million by 2018/19.
I mean, we’re talking between 15 and 30 million dollars just for the engine.
Right, but the overwhelming majority of the cost is tied up in the airframe (advanced shaping and stealth don’t come cheap) and the engine (the F-35’s engine is a feat of engineering which will live through the ages, making as much power as a lot of two-engined installations). You can do those a whole lot more cheaply for a drone, and if you invest in better missiles (the Meteor outclasses the AMRAAM-D in every way nowadays), you can make up the survivability deficit with missile improvements and networked combatants. 1:1 loss ratios drones vs. advanced fighters are impossible, but 3-4:1 seems feasible to me both technically and financially.
I don’t think that’s the way the world will go, personally: the primacy of EW is the trend to bet on, in my considered opinion, and removing the man in the loop seems very unlikely to meet broad political acceptance. That said, there’s definitely room for asymmetry in air warfare, and we discount that at our peril.
I agree that in any sane world millions of dollars for anything “disposable” is ludicrous. But when you consider the cost of keeping a full F-35 squadron operational, you have gone beyond ludicrous into total insanity. Putting aside maintenance costs entirely you currently need 36 F-35s to keep a 12-plane squadron in the air because at any given moment they are mostly grounded due to one fault or another. If it’s a carrier squadron and not ground-based that means a single working squadron costs over $1 billion just in acquisition. Even if a good drone costs $50 million, so long as it is reliable you can field over 200 of them or an entire air group for the price of that one F-35 squadron.
But I still don’t believe drones will replace piloted aircraft anyway for combat against technically advanced foes until autonomous AI is allowed on the battlefield.