Hacker News Reader

Comment by Animats | Hacker News Reader

The article has a good intro, and then never goes on to discuss the "key strengths". It's just clickbait. They're also late to the party on this. The US professional military has been discussing this for over a decade.

There have already been more than a half dozen articles on drones in Parameters, the U.S. Army War College's journal. "The coercive logic of militant drone use" [1] is worth reading. This covers how non-state groups have used drones, what worked and what didn't. The US Army is unused to fighting under a hostile sky, and now, a few times, they've had to. "In 2017, then commander of US Special Operations Command General Raymond A. Thomas noted, “[the] most daunting problem [of 2016] was an adaptive enemy who, for a time, enjoyed tactical superiority in the airspace under our conventional air superiority in the form of commercially available drones and fuel-expedient (?) weapons systems, and our only available response was small arms fire.” They also mention a drone disabling an Israeli naval vessel in 2006.

There's been a lot more military use of drones than is generally realized. Much of it is for recon. Commercial hobbyist drones are more than good enough for finding out what the enemy is up to, and most drone use by militant groups is for that.

So far, that article indicates, drones have been unable to yield a decisive advantage in any conflict. At least to militant groups.

Not many people read Parameters. That paper has only 65 downloads. More should.

[1] https://press.armywarcollege.edu/cgi/viewcontent.cgi?article...

View on HN

>So far, that article indicates, drones have been unable to yield a decisive advantage in any conflict

Drones gave Azerbaijan a decisive advantage in the recent conflict with Armenia: https://www.csis.org/analysis/air-and-missile-war-nagorno-ka...

Only once. Everyone has paid attention, and the next drone attacks will be suitably defended against. The fact is the "killer bot" scenario is just a one time thing. After it happens, all cities or armies or whatever, will have very solid defences installed against swarming or even just loitering bots. Flying drones are incredibly fragile and exposed. They are slow and easy to kill.

>Only once. Everyone has paid attention, and the next drone attacks will be suitably defended against.

no. Azerbajan/Turkey went into it only after verifying that it works, to the devastating effect, against the Russian made air-defenses in Syria and Libya. Now Ukraine has bought these drones and is starting their production. In the next year or two we'll see them being used to the same great effect in Donbass. So far Russia hasn't managed to come up with any meaningful defense. If anything the recognition of the issue has only started to dawn on them, and by the time they manage something the Turkish/Ukranian (they discuss collaborative development) drones will improve another level up - drone capabilities at the current stage (beginning of the hockey stick) are much faster to improve - add speed, a bit of stealth, [partial] autonomy, coordinated swarming. It is a new paradigm shifting generation of warfare like iron navy ships, machine guns, planes, missiles, etc. were in their respective times.

Rafael Systems sells Drone Dome anti drone mobile defense.

I think they even proved themselves in the recent Gaza campaign.

I don't think there were any real military drones. While an RC model/quad with a grenade does qualify as a weapon, the purposely built military drones have orders of magnitude better communication over longer range, incl. remote high precision targeting, loitering times, and carry a high precision stand-off missiles (that is an important point because the defense has to shoot the small fast incoming missile while the drone itself carries several missiles and loiters outside of the CIWS/SHORAD range waiting for the window of opportunity, and thus shooting the drone itself requires much more expensive/heavy air-defense system).

The iron dome defends against saturation attacks. You see the average HN poster come up with swarm buzzwords and then talk about how defense against saturation attacks is impossible meanwhile the iron dome intercepts 90%+ of hamas rockets even though they are sending thousands of them.

They don't even claim it works on bigger military drones, but smaller drones or drone swarms. It uses a microwave beam to disable or a high energy laser to shoot down the drones. I think the bigger ones like Predator could be shot down by other drones armed with low payload missiles. Repurposed Iron Dome style defenses could also be used at shorter range.

Slow and easy to kill? Seen a drone race? At what range and with what weapon would they be easy to kill? How long would the kill take? At what range could the drone be killed at?

Sure often drones fly in straight lines or hover. But in an adversarial environment they could easily move pseudo randomly and be quite difficult to kill.

And even drone races are slow and predictable compared to what is possible with an applied military mind and decent tools. I suspect (or hope) we already have some decent options in the top secret space.

Nothing we have now (publicly) is going to do a damn against a shipping container full of 2-5 lb drones each carrying 1 lb of C4 in a swarm with cheap off the shelf sensors going 100 mph nap of the earth in a distributed pattern. And that is currently entirely doable with off the shelf drones using existing software and published techniques for peanuts. Add in some decent programming and logistics talent, and that is a huge problem.

>Nothing we have now (publicly) is going to do a damn against a shipping container full of 2-5 lb drones each carrying 1 lb of C4 in a swarm with cheap off the shelf sensors going 100 mph nap of the earth in a distributed pattern. And that is currently entirely doable with off the shelf drones using existing software and published techniques for peanuts. Add in some decent programming and logistics talent, and that is a huge problem.

Existing technology can take down conventional drones through Electronic Warfare and EW scales to an infinite number of drones assuming each drone is vulnerable.

That is not true. Existing technology can take out remote controlled conventional drones with EW.

I know of no existing (human proximity safe) EW that can handle optical based or internal (IMU) inertial autonomous flight, and both of those are more than good enough in this scenario (though optical may limit the upper speed bounds depending on the onboard processing capabilities). IMU based (pre programmed) could easily allow no active sensor mode, though would require pre-scouting the route.

The Indirect Fires Protection Capability (IFPC) is probably what you're looking for.

The US Army is looking to field a demonstrator by 2022 & deploy by 2024.

They're exploring both laser and microwave / EM for the counter-UAS role. In complement to existing SAMs for higher-value/capability targets.

https://asc.army.mil/web/portfolio-item/ms-ifpc_inc_2-i/

Thank god something is in the pipeline! Thank you for the links.

If you mass produce these things you are going to leave traces of your logistics supply chain. Even if you brought them into a staging are one at a time your frequent trips would lead back to where they were made or by whom.

Which wouldn’t stop someone using it in a declared war - which we haven’t had yet by an actor likely to want it. Doesn’t mean it isn’t sitting in a military port somewhere though.

And since China makes like 99% of the parts used in drones, they very well could already have a couple hundred of those lying around and we’d never know.

At some point 3d printing will make localized offgrid drone production entirely feasible.

The chips, PCB, and all electronics have their embedded serial number in them. It's very likely many of these will survive a direct explosion, and once these identifiers are extracted from them, the supplier/manufacturer/vendor/customer will be identified using standard investigation methods. So while there's little possibility of some unknown Unabomber-style decade long rampage using this tech, they could be very destructive in the short-term.

My nightmare scenario is someone flying hundreds of these in formation on a stadium full of people. They could be launched from a parking lot from a single van or pickup truck. If you sacrifice the ability to hover by giving them wings, you might get very high range and not have to risk yourself at all while launching a devastating attack.

I'm sort of surprised we haven't seen this already. It's not even that hard or expensive to build.

This made me realize something like a TAS drone race would be interesting to watch.

What is TAS? Haven’t heard that term before. total autonomous system?

Tool-assisted speedruns. People use emulators and record inputs frame-by-frame, allowing nearly perfect performance and even reliable exploitation of game vulnerabilities. I remember one where the author exploited a bug in such a way that it completely reprogrammed the game:

http://tasvideos.org/5384S.html

Check it out, it's incredibly fun:

http://tasvideos.org/

There's an entire category for movies containing arbitrary code execution:

http://tasvideos.org/Movies-C3050Y.html

"tool assisted speedrun" is a term in video game speedruns referring to the use of software/external tools to help with speedrunning games. So imagine a drone race where everything is programmed rather than remote controlled by a pilot

Modern CIWS are designed to be effective against substantially faster and more agile rockets. Drones really are slow and easy to take down compared to guided rockets and especially kinetic ones, and the military has been working on means of defending against those for decades. The Phalanx is being modified to serve as an anti-UAS system, for example. The challenge is more to do with detection than firing.

Dunno, do agile rockets dodge? Or just follow a simple preprogrammed patterns (like low, then pop up to attack from above)?

Not sure a CIWS system that can hit a mach 2 missile that doesn't dodge would necessarily handle the same $$$ spent on a N cheaper drones. How many minutes of firing do modern CIWS systems typically have?

Loitering at max engagement range and dodging might well end up using ammo at a rate that the CIWS can not sustain.

Also I suspect any CIWS can get a kill on just about any part of a missile, where drones often have very small kill zones (often battery or gas, payload, motor(s), and control) that might well be just a few % of the size of the drone.

I googled the phalanx, looks like it typically has a drum of ammo that lasts for about 3 seconds. Is that 3 drones worth? 10? 100?

How do you "dodge" airburst ammunition? Can your drone instantaneously change direction at the speed of an explosion? What the hell am I reading here?

https://www.youtube.com/watch?v=bdwjcayPuag

Edit: Look they even made a smaller system! https://youtu.be/Drm3kdDYBI4

This seems to aim by calculating the trajectory of the target. How will it fare if the target is flown in first person perspective by a human in an unpredictable way?

It’s airburst ammunition which can be fired faster and travels faster than a human can react. The airburst is timed to trigger at the calculated intersection of trajectories (or somewhere in that neighbourhood), significantly improving your chances of scoring a hit even with a high circular error probable.

> Also I suspect any CIWS can get a kill on just about any part of a missile, where drones often have very small kill zones (often battery or gas, payload, motor(s), and control) that might well be just a few % of the size of the drone.

I guarantee you that just grazing a drone with a 20mm round will still kill it, even if you don't "hit" anything critical. :P

Really? Say you blow a 1" hole in a wing that's got some balsa struts and some thin fabric? Or a wing just made out of a sheet of foam? Both are common on cheap drones.

Sure the 20mm round represents an amazing amount of kinetic energy, but the drone doesn't offer any resistance, and I wouldn't expect the round to even notice. Sure the disturbance in the air might throw the drone for a few loops, but any decent software will recover.

Sure hitting more substantial targets the target itself becomes of the shrapnel that causes an ever increasing zone of damage. Not clear to me that a piece of paper or thin fabric would exhibit the same behavior.

Speaking as someone who's built and destroyed lots of homebuilt aircraft, and shot lots of things with guns of various sizes...

1. I think you're conflating "cheap agile drone" with "foam or balsa+durakote fixed wing aircraft".

2. Probable engagement aspect for a drone is not "straight up", as would be necessary for your scenario.

3. A little itty bitty hole in a wing does a whole lot more than you'd think.

4. Expecting a high velocity cannon round to just punch a clean hole in the wing is a little much.

Foam's a bit more robust, but a small pebble at a few miles per hour has wrecked my whole durakote day before.

And yes, there's degenerate cases, like you could have a whole sheet of dollar tree foam board maybe take a hit or two.

In the C-RAM role, Phalanx uses 20mm M940 MPT-SD rounds.

Explosive (self-destructing) and incendiary

https://www.gd-ots.com/munitions/medium-caliber-ammunition/2...

> Dunno, do agile rockets dodge? Or just follow a simple preprogrammed patterns (like low, then pop up to attack from above)?

Modern antiship missiles have complex programs. Some are capable of flying in formation with each other, with one missile flying high for visibility and communicating with the other missiles in the formation that are flying very low. If/when the missile on top is destroyed, another missile in the formation can be automatically selected to take its place.

Most of those systems are designed to target dozens of incoming targets. It’s entirely reasonable and plausible for drone swarms to number in the thousands of targets.

That ciws will likely exhaust its magazine before it can take out all the drones unless I’m missing something.

Lasers. You just have to put "laser defense drones" into Google to get a whole slew of completed or in development solutions. Also EMPS, counter defensive swarms of drones. The list goes on...

Here is the system I am aware of:

https://youtu.be/1DXpPmpmcak?t=365 (watch this entire video for air defense against saturation attacks)

https://www.youtube.com/watch?v=PV3jfR-FUFc

That's why Israel is begging the US for another billion dollars for reloads for their Iron Dome system. They're using expensive guided missiles to shoot down cheap unguided rockets.

Deploying point defense technologies over a city/civilian targets has some obvious problems, though.

Lasers, electronic countermeasures, tangly stuff, etc, are much more attractive than systems that fire thousands of cannon rounds per minute or that launch explosive missiles.

From the phalanx wiki:

Whereas naval Phalanx systems fire tungsten armor-piercing rounds, the C-RAM uses the 20–mm HEIT-SD (High-Explosive Incendiary Tracer, Self-Destruct) ammunition, originally developed for the M163 Vulcan Air Defense System.[26][34] These rounds explode on impact with the target, or on tracer burnout, thereby greatly reducing the risk of collateral damage from rounds that fail to hit their target.

Not that the risk is zero, but at least it's "greatly reduced".

You wouldn't generally let a C-RAM fire at a low angle over a city. And even if you did, the problem of how many of these you need for decent coverage becomes large, because there's no ideal placement that doesn't involve shooting straight at friendly things from some approach vectors.

A CIWS is not a reasonable way to stop drone attacks on civilian targets.

And one of tue whole advantages of a drone army is that it can fly between zero and ten feet off the ground - thereby undercutting lots of defenses that are designed to point in the air.

Flying nape-of-the-earth is a hard counter to CIWS-style defense systems. For antiship missiles this tactic doesn't help a whole lot, but for drones over land? C-RAM can't see through a hill, let alone shoot through one.

> Modern CIWS are designed to be effective against substantially faster and more agile rockets.

The thing is exactly that those drones are so slow, and low.

Phalanx performance against sea skimming missiles is know to be poor.

And that on dielectric, flat sea surface.

Even military drones weighting few tons can fly autonomously under 100m.

Ironically, it's now depressions in the terrain which will provide more air defence since they will force terrain following drones, and cruise missiles to move at oblique angles.

Basically the problem is the same as with low, and slow cruise missiles, except worse.

Does it at all change the equation that you could trivially fly the drone 2 feet above the ground?

My university has been working with the army to engineer drone vs human engagements. They originally set out to increase the drones' efficacy against humans wielding small arms (shotguns/etc), but apparently found out that was trivial. The drones very, very quickly overwhelmingly surpass the ability of humans to engage and defeat the drones.

The research pivoted very early to "training humans to more successfully defend against small drone attacks". I'm not sure how much success they've had or what success is defined as for something like this.

Here's a line of cops shooting at race drones. They would've had plenty of time to escape if they weren't sticking around to find out how long it takes to get shot down. These were race drones from 5 years ago and modern ones can be around twice as fast. Not to mention you don't want to dump a magazine upwards in your own territory.

https://www.youtube.com/watch?v=xq0oCM37oZA

> Not to mention you don't want to dump a magazine upwards in your own territory.

That is what birdshot is for. Birdshot, e.g. "No. 9 or 10" shot, are tiny lead BBs about 2mm diameter. Because they are small, they have a large surface area to mass ratio, and consequently, they fall back down to earth at a low, safe speed. When they land on dry leaves, they bounce off and sound a bit like rain. When it is shot into the air and falls back down onto you, it won't even bruise you, let alone injure you. (However it can kill if fired directly at somebody.)

How about a super soaker instead of a pistol? Especially one of the CPS soakers that got banned https://www.youtube.com/watch?v=resgaF7MeR4

Fun video. Definitely showed how easy it is to shoot down drones of the consumer variety.

If you can hit it with a pistol it’s easy to hit.

Current point defense systems are already capable of placing a 20mm round with a muzzle velocity of 1000m/s onto a falling mortar shell in the last three seconds of its flight time. How hard do you think drones are by comparison?

It depends on how long it takes the weapon to lock on and align. If it takes 1s to work out the trajectory and fire, the drone just needs to change its trajectory every second.

But I guess taking down a drone is not like taking out a mortar shell. It is probably much easier with some sort of flak or even launching a net.

> How hard do you think drones are by comparison?

Very. The calculation made for each round could be irrelevant before they travel 500m. Drones also have no obligation to approach at the same convenient high angle mortars do. 2-3 drones weaving between buildings faster than a human can aim? Sounds terrifying.

How would they do if the mortar shell engaged in rapid, unpredictable evasive maneuvers at low altitude behind cover most of the way?

So what you're suggesting is instead flying in a giant predictable parabolic flight pattern, just dropping the drone in front of the actual targets and let the big cannon do the work? Friendly-fire/own-goal for the win.

ballistic trajectory versus active evasive behavior.

https://apps.dtic.mil/dtic/tr/fulltext/u2/a426717.pdf

A target with a 300mm frontal diameter, capable of evading at 10g, is expected to require 6 rounds to achieve a P_hit of 0.5 at 200m range, based on the parameters of your typical Phalanx CIWS. Said Phalanx weapon system might fire a burst of 1s duration in such an engagement, which is 75 rounds.

For a missile engagement, that's a problematic number, but for a drone that's probably only moving a few dozen m/s and carrying a small payload, it's an entirely satisfactory engagement.

Phalanx is an early 1980s weapon system. The Army is going to be fielding directed-energy SHORAD systems next year, equipped with 50kW lasers.

Phalanx usually has about 1500 rounds ready, so 20 drones at 75 rounds each will run it out. Phalanx is large, heavy, and costly, and can't be readily be used in built up areas at low angles.

A few dozen m/s is probably a good estimate for quadcopter style drones, but low hundreds m/s should be easy for fixed wing drones, so at 200 meter engagement range (that paper makes it pretty clear that maneuvering targets are hard beyond that) there is only time to hit a few drones before they arrive.

So maybe drone swarms will work even against current CIWS.

>So maybe drone swarms will work even against current CIWS.

Sure, but current drone swarms wouldn't.

Exactly - mortars predictable trajectory against a background free of ground clutter make them easy to predict the future location of.

A drone optimally randomly dodging at treetop height is a much more difficult target.

Why the hell would you dodge at tree top height? That makes absolutely no sense. Just fly a few centimeters off the ground without dodging and no air defense is going to hit you in a civilian area.

Or just fly near the ground so that air defense will kill the target of the drone...

They could fly low: the tech isn't fully there yet but skydio drones show the potential.

Actually, quite hard. Mortar shells don't suddenly change direction.

> Current point defense systems

Which are available to how many militaries?

They are slow and easy to kill compared to fighter jets, the grand parent is not talking about hobbyist drones but UCAVS.

what does a done race have to do with a loitering munition ?

have you seen a harop? its huge,its built to fly slow and for a long while and carry big payload and has zero evasion capability

https://www.theguardian.com/world/2010/jul/25/reaper-drone-m...

Not all drones are AI, there are remote drones that are very fast and very hard to kill, pulling Gforces that would make normal pilots pass out. With the US declaration that drones are "munitions", we have effectively stated that we can launch them into the airspace of countries that we aren't technically at war with, and claim it was "border fire", like we did with Afghanistan.

The primary defense for the time being is 'hide.' US Marine tacticians, for example, are shifting priority from hardening an impromptu position (ie a foxhole) to concealing it.

Check out some of the Azerbaijani drone strike footage with IR on Youtube. They'll need to go deep and avoid any above ground movement and it better not look like a bunker. Unless effective drone defense (other drones?) is employed using cheap drones are a game changer.

They are very small, and high up. As the AI gets better they will also become better at evading fire, just as mosquitoes are really hard to get. Try catching a cat that doesn't want to be caught. Equip the cat with an explosive warhead and make it a thousand cat swarm.

> Everyone has paid attention, and the next drone attacks will be suitably defended against

How? They fly just few hundred meters above the ground, launch from field runways, and for the price of a single B-2, a country can buy 2000 of them.

Jamming? Even a country like Turkey was able to devise a passable inertial navigation, and and programming for autonomous operation way exceeding an easy to trick "attack anything that moves" mode.

To counter them, you will need a very sizeable air force fitted with best radars possible to track them down individually, and strafe with a cannon (since no missile currently can lock onto something so small, slow, cold, and low flying.)

> To counter them, you will need a very sizeable air force fitted with best radars possible to track them down individually, and strafe with a cannon

Or, you know, currently fielded SHORAD systems and CIWS?

What are their range, and height of radar horizon?

Even if that wouldn't be a problem, you will still need really a lot of that.

On the Phalanx, about 1.5 - 5.5 km?

Either self-radar or tied into a broader net.

That's current-gen, repurposed from something designed against cruise missiles.

Next gen will probably optimize for or supplement with lighter weight projectiles, depending on whether lighter or heavier drones proliferate.

What about their speed and range? If they fly fast they're not going far. If they go far they're slow moving targets. And if they go fast and far they have expensive rockets attached and you're being hit with missiles instead.

The biggest threat today are tactical drones from just below a ton to a few tons.

100-300km - way outside of range of artillery. And they are all field runway launchable, so you don't really get anything from bombing the airfield.

And most of them are made to fly below heights at which any air defence is effective.

> They are slow and easy to kill.

The drones flying slow, and low makes them harder to kill, not easier.

> commercially available drones and fuel-expedient (?)

A transcription error that has been carried forward over several sources, apparently. The general undoubtedly said "field-expedient", meaning devices that can be cobbled together in the field. Take a commercial drone from ToysRUs and tape a hand grenade to it, and you have a field-expedient war drone.

Tape a knife to a roomba and you have a weapon of war.

The key strength of drones is that they kill who you want to kill, or destroy the materiel you want to destroy, without any collateral damage at all. Coupled with adequate intelligence, which they also help obtain, drones make nuclear weapons obsolete; they are a dramatically more effective means of coercion. There are a lot of fantasies in this thread about how they're easy to defend against, and that would be true if we were talking about drones that cost the same as a fighter jet or a Tomahawk or even a Toyota, but you can mass-produce drones that are capable of delivering sub-kilogram munitions with centimeter precision for under ten dollars.

So the relevant comparison isn't a slow cheap drone versus a Tomahawk or a Toyota; it's a hundred thousand slow cheap drones versus a Tomahawk, or two thousand slow cheap drones versus a Toyota. A hundred thousand slow cheap drones is enough to coerce all the officers of any army into resigning; all you need is the realtime whereabouts of their parents, siblings, children, and spouses, and effective ransomware that reliably spares the children of officers who have already resigned, or who haven't yet been contacted. And no CIWS is going to help with that, even if it doesn't run out of ammo after the first fifteen drones, because the children aren't on an aircraft carrier or an army base.

Jef Raskin pointed most of this out in 02001 in an underrated essay, 'next time it can be worse'. At least so far, it hasn't been.

This is a tech power fantasy. Its not at all representative of the state of the art and disregards a lot of difficulties.

Massed drones absolutely can wreck terrible destruction, death by a thousand cuts to any facility or base or airfield. But the kind of amassed yet precise strike against a hundred targeted individuals, with smart targeting and somehow knowledge of their career status?! Come on.

I wouldn't even finish that movie, let alone fund that proposal.

smart targeting is clearly crucial, yes. but the objective of warfare is not to wreak (or wreck, heh) terrible destruction; it is to break the enemy's will to resist. by themselves, drones don't tell you who to target or where they are; they just gather surveillance data and deliver munitions, very precisely and in a way that is very difficult to stop. in combination with things like facebook data and cellphone subscriber data, that's an extremely powerful capability

keep in mind that what is happening today is not what will happen tomorrow or what can happen in five years

you are rejecting the possibility because of its unfamiliarity, much as the navy rejected the possibility of a surprise attack on pearl harbor

I'm challenging the supporting assumptions you described. Real time career knowledge, real time non visual person tracking, real time coordinated search across cities, real time IFF, lethal payloads on them all, real time localization. There's a lot of requirements there. An alien invasion would probably use clouds of drones in this way. The US could. Russia, China. After some testing. But not the next ISIS. And those states that can already have better options for targeting at range on a battlefield. And I'm challenging the use case. Are they within US borders? How? Flew in? Launched from within? How did they get close enough? A ship? Infiltration?

Pearl harbor was easy, required no new tech, just will. And Pearl Harbor was well known before the attack, had been demonstrated years before. This isn't that.

> Real time career knowledge, real time non visual person tracking, real time coordinated search across cities, real time IFF, lethal payloads on them all, real time localization. There's a lot of requirements there. ... Are they within US borders? How? Flew in? Launched from within? How did they get close enough? A ship? Infiltration?

Rather than go through and carefully solve the fascinating puzzle of how to destroy civilization step by step, I'm just going to point out that these things are not actually very difficult. Instead of hoping that the next Daesh can't recruit anyone who knows how to build an IMSI catcher, write a Facebook app, or use LinkedIn (or VKontakte if their adversary is Russia), think about what happens when there's a conflict between two adversaries that can.

> The US could. Russia, China. After some testing. But not the next ISIS.

Certainly the three states you mention could develop that capability, yes, although they don't currently have Slaughterbots fleets ready to launch. And Daesh certainly wasn't able to develop it, although they did drop a bunch of unguided hand grenades on people from remotely-piloted quadcopters. But what about Turkey? France? Rebels in the West Bank less incompetent than Hamas? The Zetas, who as I understand it have been running their own cellphone networks for 10 years? The Sinaloa cartel, which has thousands of drones carrying contraband across the US border? The Jalisco cartel, which is already targeting the families of police officers? The US-backed Guaidó faction in Venezuela? The US-backed military dictatorship currently running Egypt?

This isn't something that requires the budget of an aircraft carrier, and it's not something you can see from a satellite before it happens.

Sure. 'Think of the possibilities' is a fun exercise. I'm more in the 'will it scale' camp, that's all.

I've actually worked in this area before. There's more attention on this than you might imagine, and less than it deserves. There's also better ways to use drones to cause terror and disrupt operations that I obv won't get into. Maybe its not tech fantasy as I originally said, but it is much further out than you originally said as well, and it's a bit overkill. If you wanted to have the effect of causing fear in soldiers using drones, I think (but don't know) there are easier ways that don't require the same limiting assumptuons about technology, logistics and scaling.

So, what do you think we can do to prevent it?

Nice try, China.

Heh! Actually, although I'm in Argentina, I thoroughly endorse 墨子's defensive program: if every country is safe against attacks, then all the people will be safe, while if every country is vulnerable to attacks, then life will be bitter and filled with grief. So, just as 墨子's followers worked to improve the fortifications of the various states of China so that none could gain by attacking another, I think we should figure out how to reduce our collective vulnerability to this kind of drone-swarm attack.

Except that I think it's kind of impossible and we are basically doomed.

Minus the emotional wording, this has been the justification for large standing militaries in every nation state ever. I mean, it's the Dept of Defense here in the US.

Right, and the US isn't even a nation-state, despite the persistent efforts of certain reactionary groups to change that. But it's a backwards justification, because thousands of years of history show us that large standing militaries increase the risk of war, while fortifications diminish it.

$10 will get you a palm sized drone with a flight time measured in minutes. Stop dreaming. The type of drone that people are scared of costs at least $1000.

Hamas sends thousands of rockets in that budget range and nobody is scared that rocket swarms are going to kill thousands of Israelis. IDF retaliation killed more innocent people in a single bombing than all Hamas rockets combined.

> $10 will get you a palm sized drone with a flight time measured in minutes.

retail, and battery-powered. what will a $10 bom get you with a nitromethane-powered engine or a glider dropped from a helium balloon

> The type of drone that people are scared of costs at least $1000.

they need to rethink that. stop fighting the last war

> Hamas sends thousands of rockets in that budget range and nobody is scared that rocket swarms are going to kill thousands of Israelis

that's because hamas is too dumb to put glider wings and ardupilot on each rocket and possibly too deprived of facebook and imei-logger data to know where to find the specific thousands of israelis who are family members of the invading officers

'our enemy might be as incompetent as the hamas model rocketry club' is not a great way to think out how to provide security

this is not a dream. this is a nightmare

listen, i know this scenario isn't happening today. but it's going to happen within the next few years, and the fact that it isn't happening today isn't much of an argument against that

Who is going to try that and why wouldn't they meet the business end of a nuclear missile?

possibly because the officers in the nuclear silo want some of their family members to still be alive when they go home. or possibly because the nuclear dudes don't know who to nuke. or possibly at the point where your answer to a military threat is 'well if they try that we could just nuke them' you are kind of admitting that it is a very serious threat indeed; note that nobody got nuked in response to 9/11

That paper has only 65 downloads.

Now. 640 downloads. Still not many, but a huge number for Parameters, a publication read by more real generals than armchair generals.

Reply on news.ycombinator.com