Anti-Drone defenses?

[DataPacRat]

I've recently been reminded of the video, Slaughterbots (NSFW, violent imagery); and it's relevant to a near-future setting I'm detailing.

How would such bots best be described in GURPS terms, and what sorts of defenses could be feasible against them?

My first guess would be to start with the Flier Robobugs from p111 of Ultra-Tech; "3 grams of shaped-charge explosives" is two-thirds of a robobug's 0.01 lb weaponry, so it seems feasible, though I'm not quite sure how to stat out the explosive charge itself. Do I need to drag out my 3e Vehicles book?


As for responses, I can think of some of the obvious first-level defenses at various stages: discreetly-armored hats, trying to program some anti-drone drones, concealing one's face and other biometrics when in public, scrubbing one's social media both to minimize harvestable biometrics and to avoid becoming a target, and so on; but I'm still pondering how the various attackers would respond to those steps, let alone what the responses to that level would be.

What ideas come to your mind out of all of this?

[starslayer]

I don't think these fellows have enough on-board AI to do the facial recognition, so they need to 'phone home' for their 'brain' to work.

Thus all you need to do is jam a few simple radio frequencies and all the bot will be able to do is hover (or possibly drop right to the ground if it does not have enough smarts to hover).

A flyswatter or butterfly net will take care of the rest as of that point.

[DataPacRat]

Originally Posted by starslayer

I don't think these fellows have enough on-board AI to do the facial recognition, so they need to 'phone home' for their 'brain' to work.

At standard TL9, the robobugs' tiny computers are Complexity 3. The setting I'm working on has roughly TL10 computers, so that'd be Complexity 4 or 5. 4e's Ultra-Tech doesn't even seem to apply a Complexity for augmented-reality-style facial recognition. The TL9 computer implant, with a tiny computer, on pUT215, says that such hardware is powerful enough to include an 'optical recognition' feature.

Thus all you need to do is jam a few simple radio frequencies and all the bot will be able to do is hover (or possibly drop right to the ground if it does not have enough smarts to hover).

A flyswatter or butterfly net will take care of the rest as of that point.

That seems to... significantly underestimate the creativity that could be applied by potential drone-launching groups. I can think of several potential solutions with just a few seconds of thought; they'd have lots more incentive to figure out which solutions would actually get the job done.

[Refplace]

They may have problems with facial recognition with onboard computers but they should still have enough capability for default programs.
And a queen drone thats a bit bigger could identify targets.

[Extrarius]

Originally Posted by starslayer

[...]Thus all you need to do is jam a few simple radio frequencies and all the bot will be able to do is hover (or possibly drop right to the ground if it does not have enough smarts to hover).[...]

With modern spread-spectrum frequency hopping techniques, it is very difficult to jam a robust signal without jamming everything that uses radio waves at all. With future advancements, I wouldn't be surprised for it to become essentially impossible to jam a robust signal without jamming literally everything else that uses radio frequencies. In fact, the jamming signal might have to be strong enough to interfere with devices that interact strongly with EM waves (like microphones and speakers that could be used in wired communicators) even when they aren't designed to be radio transmitters or radio receivers.

It also wouldn't be hard to make a drone detect jamming (successful or not) and switch to homing in on the strongest nearby radio emitter.

[Fred Brackin]

Originally Posted by DataPacRat

I'

My first guess would be to start with the Flier Robobugs from p111 of Ultra-Tech; "3 grams of shaped-charge explosives" is two-thirds of a robobug's 0.01 lb weaponry, so it seems feasible, though I'm not quite sure how to stat out the explosive charge itself. Do I need to drag out my 3e Vehicles book?

No, Campaigns covers simple explosive calculations. You do need to convert from metric though.

3 grams appears to be0.0066 lbs and multiply that by the REF of 1.4 for C4.. You take the square root of that and get 0.096. Multiply that by the 42 pts of damage (6Dx2) you'd get from a lb of TNT and it comes out to 4.0 which would round down to 1D of explosive concussion. Making a shaped charge probably wouldn't be worth the trouble.

Even going to the top explosive from HT (CL20 I think) and you might get up to 5 pts.

[thrash]

Originally Posted by Fred Brackin

Making a shaped charge probably wouldn't be worth the trouble.

Unless it's programmed to find the victim's mouth, eyes, or ears.

[Fred Brackin]

Originally Posted by thrash

Unless it's programmed to find the victim's mouth, eyes, or ears.

It would probably be more effective to plant the charge directly agaisnt the skull next to the brainstem. That's the part you want to hit. A shaped charge jet so small miight not reach all the way from the eye to the back of the skull.

If you can be that precise go for the carotid artery instead with a sharp blade and 3 grams of anti-coagulant. Lethal doses of botulin or ricin toxins are also very small.

[DataPacRat]

Originally Posted by Fred Brackin

It would probably be more effective to plant the charge directly agaisnt the skull next to the brainstem. That's the part you want to hit. A shaped charge jet so small miight not reach all the way from the eye to the back of the skull.

If you can be that precise go for the carotid artery instead with a sharp blade and 3 grams of anti-coagulant. Lethal doses of botulin or ricin toxins are also very small.

In the video, during the presentation, there's a split-second image showing half-a-dozen different drones: two simple anti-personnel quadcopters, one wall/window breacher, and what looks like a dragonfly-style ornithopter with a hypodermic needle strapped to its back. (And two others.)

(Another split-second image implies that $2.5M can buy 700,000 tiny drones, implying that at least one model costs under $4...)

[cptbutton]

And old SF reference to the concept: The Hellhound Project by Ron Goulart.