Comment by ryanmarsh | Hacker News Reader

Comment by ryanmarsh | Hacker News Reader

Parent Full thread

once robots get more capable

There’s a common misconception that robots are not yet capable enough. This obscures the present, unexecuted, reality.

It is well within the budget of many nations to build grenade carrying swarms of drones capable of autonomously patrolling and attacking human targets. The fact that we have not seen this on a wide scale has no bearing on present capabilities, or that which is within reach. If you search YouTube you can already find examples of ISIS using remote piloted inexpensive drones to attack armored targets with devastating effectiveness.

This is here, now.

As a former light infantry grunt I have a visceral feel for the breathtaking transformation of the battle space we may soon witness. Within an AO it is very challenging to deny access to light infantry, even in the most foreboding terrain. Autonomous drones can and will effectively deny infantry access to large areas. This fundamentally changes ground warfare, even guerrilla.

Use logic. Can a drone recognize someone holding a weapon? Can a system be devised to demark areas of friendlies? Are all nations as cautious as the US when it comes to each weapons release?

Agreed. Modern commercially available drones can follow targets, even if they are temporarily obscured, detect faces, path around obstacles and perform preprogrammed tasks. The higher end ones typically carry enough compute power to do fairly advanced video recognition and execute logic. All that’s really missing for weaponization is an articulated firearm.

What do you need an articulated firearm for? Isn't some C4 good enough?

Hell, why do you need the drone to be autonomous? There are $100 drones that can be piloted remotely from kilometers away.

Also, you can leave scores of drones lying in wait, and alert a small number of pilots to fly the right ones when motion is detected.

> Isn't some C4 good enough?

Not by itself. An area weapon is fine, which is why I said grenade. They are effective even with enemy in defilade. Rifles are cumbersome, give their position away and have recoil.

> why do you need the drone to be autonomous

Electronic Countermeasures

This is why autonomy might be desirable:

https://duckduckgo.com/?q=turkish+soldiers+drone+jammer&t=ff...

We are getting closer to the cyberpunk squad of guys with cf body armor, one guy with a drone jammer and a couple more with high suppressible firepower.

What is AO?

Area of operations

https://www.quora.com/What-does-AO-mean-in-the-military

1500 rounds in a Phalanx, so need 1510 drones (catapults, radar, any aircraft on deck), at, say $5000 each. $7.5m to take a carrier out of the battle. I hope the Navy has wargamed this or demonstrated the phalanx software wont shit itself with that many targets.

Check out the US Navy’s new laser cannon!

It's wild to imagine but 100% believable to think that the US Navy already has technology to defeat a large mass of cheap weaponized drones. And even more entertaining if it's lasers.

It would be a shame if those drones were covered with mirrors,wouldn't it?

hopefully the navy isn't superstitious.

That means more weight, more cost, for something that is ultimately disposable.

Something like a CIWS could work but it’s big and heavy and probably would need some time to calibrate. So you can’t have them with you everywhere, especially on a foot patrol, raid, recon, etc...

How are the drones different from a simple bomb or missile then? You might as well just build a smaller SDB with more strikes. The vast majority of these drone weapons boil down to the concept of a guided bomb. Drones have a limited operational time measured in minutes and they can only attack once. The only difference to conventional bombs is that smaller explosives cause less collateral damage.

Well, once someone has 1510 drones capable of taking a carrier out of battle, they'll just toss a few dozen smaller turrets on the carrier that fire birdshot at full auto.

Check out this wargame, where floats of small inexpensive boats dominate a traditional navy:

https://en.wikipedia.org/wiki/Millennium_Challenge_2002

Assuming the drones/missiles cannot be easily jammed, the main counter to such an opponent would be to:

1) Know their capabilities 2) Stay out of range of the drone/missile swarm, until you have rendered them harmless (by destroying or blinding them).

There are unjammable communications. Essentially ad-hoc networks using laser beams, coordinated from a plane.

The reason it's unjammable is that you need to be in line-of-sight of the laser between 2 devices to jam it.

Blinding swarms of drones with communications and coordination between them is impossible. If necessary, they'll just get their sight from a satellite above.

So this would fall under the assumption that they could not be jammed, in which case the trick is to stay out of range, and destroy them before launch.

Missiles like this with ranges of 100s of kilometers will be very expensive, and can be intercepted by cheaper missiles/drones (or soon, lasers).

You can jam unjammable laser. If there was something like fog or rain it could block it's line of sight.

Reply on news.ycombinator.com