Hacker News new | past | comments | ask | show | jobs | submit login
What can defend against off-the-shelf drone weapons? (dw.com)
141 points by pmoriarty on Aug 7, 2018 | hide | past | favorite | 187 comments



s/drone/guerilla tactics/ and the answer, speaking as a veteran, is nothing.

I expect to get downvoted for this, but after a tour in Iraq I firmly believe theres no weapon or countermeasure thats going to be unilaterally effective at arresting the concept of guerilla warfare and terrorism except negotiation. Otherwise, you end up dividing your city into red and green zones, and pretty soon even if a terrorist attack hasnt happened in a few months, terrorism is pretty successful in making peoples lives outright miserable. smoking kills more americans than terrorism in 2017, but we still have to take our shoes off at the airport.

People just want equality, some measure of freedom, liberty, and prosperity. Show up with tanks, rig an election in favor of your candidate, and disenfranchise an entire segment of the population like we did in Iraq, and you will find people will adopt all sorts of creative and clever methods to not only sabotage your mission, but actively eliminate your high value targets youve come to rely upon for 'nation building' or whatever geopolitical pursuit du-jour is popular.

In Iraq, we called it "winning hearts and minds." doing things to fix problems like providing clean water or just a few meals made incredible progress in crushing the enemies ability to recruit anyone more sophisticated than an angry goat herder. And he quit being angry once we fixed the sidewalk to the mosque.

In maduros case, he is reviled by a sizeable majority of his population. he openly ignores major problems like starvation, he engages his people with violent tactics, and he disregards basic human rights and living conditions. Nothing, no technology on this earth, will keep you safe if your people turn on you. There is no police or military force that can endure sustained guerilla terrorist tactics indefinitely. Not even the US Military.


In counter-terrorism they call defensive measures "hardening the target" because they know it can never be completely secure as you mention.

Guerilla tactics are the evolutionary counter-balance to an overpowering force. If there's a day when a power becomes so great it can't be resisted we're all in trouble. We like it when we're resisting (Revolutionary War) but not when we're being resisted (Iraw and Vietnam). But the tactic exists because it's effective.


What is the best way to disagree with this unilaterally?


I don't even understand what this means, let alone how you could disagree with what was said.



The same way you disagree with anything unilaterally. By yourself =)


This is the correct answer.

What can defend against someone who is resourceful, determined, and has an axe to grind? Not much.

Anyone remember Killdozer?


Killdozer is a perfect example of how we CAN defend against this kind of attack: Make sure that everyone has something to lose.

If they'd just cut that guy some slack, like maybe not destroying his business, or helping to relocate, or anything that left him with a reason to keep being a productive member of society, then the outcome would have been very different. But they fucked him into the ground, until eventually he fucked back.


http://archives.durangotelegraph.com/04-06-24/mountain_excha...

> The story of the batch plant also goes back to 1992, three years after Heemeyer had moved to the area. Heemeyer bought his two acres from the Resolution Trust Corp., the federal agency set up to handle the assets of failed savings and loan institutions. He bought the two acres for $42,000 but later agreed to sell it to the Docheff family, which wanted the property for a concrete batch plant, for $250,000. They agreed, but then he wanted $375,000 and at some later point wanted a deal worth approximately $1 million. All of this was well before the rezoning proposal hit town hall.

He got a quarter of a million for land he bought for $40k. That's hardly fucked into the ground.


Sure, but it's more that he felt he had no power and wanted to invert that powerlessness.

The (afaik) plan started after they repeatedly denied him the ability to create his own path to his property, and hit him with fines etc due to a forced rezoning. The guy (apparently) tried to negotiate and got hung out to dry a bit.

Don't know the exact details of whether they were in the right to do so, but the point is the guy felt like he had his back against the wall and had nothing left to live for.

Not condoning the guys actions though.


Never heard of killdozer, thanks for sharing!


You don't need so much work to make an impact, a truck is enough.


>Anyone remember Killdozer?

Go over to ar-15.com or some prepper forums and everyone will know exactly who you're talking about, here probably not so much. This crowd doesn't tend to idolize that kind of thing.


You don't have to idolize it to understand it, and learn from it.

Same with the Unabomber. Everyone likes to make jokes, but how many of those people have read what he wrote?


I read it and I don't think it's worth reading - it struck me as the ravings of a disgruntled pseudo-intellectual.


Nothing will make you safe from cheap attacks like IED's or Drones dropping explosives. The Q is: what will make it "safe enough to operate" under the threat. The IED threat made vehicles heavy and expensive, movement restricted.

The drone threat is similar but different. Protection of a single high value locatoin like a presidential speech (Maduro situation) is quite doable. Some combination of surveillance and kinetic and electronic countermeasures. It's not foolproof but it appears to work ok-ish (there are still high value targets speaking around the world and zero drone assasinations and now nonzero failed assasinations). The problem is that it's resource intensive and asymmetric as hell. And it's still just a question of time until the first successful attack.

A more interesting question is, what does a group of soldiers do to protect from a drone threat? They don't have a fixed location and huge resources, like a presidential speech. You can't fire a hugely expensive AA weapon at an off the shelf drone. It would seem you need some kind of weapon that is light, cheap, and reliably kills an off the shelf drone up to max altitude, and can be re-used or carried in large numbers by a single rifle squad.

It could be as simple as another drone? This is at least my guess. The most reliable and cheap thing to use to attack a drone is another drone. That would then be effectively the start of an arms race of drones. Higher, faster, better armed, better countermeasures. We'd see a replay of the evolution of fighter planes, but for plastic UAV's?


I would add to your list, "cheaper". Any drone that can be swarmed and taken down by cheaper drones is probably a bad investment.


> Show up with tanks, rig an election in favor of your candidate, and disenfranchise an entire segment of the population...

I protested the Iraq invasion in 2003 but this is a revisionist and historically inaccurate version of why it failed so spectacularly.

The U.S. didn't occupy a functional country and then destroy it. The outbreak of mass murder and tit-for-tat violence wasn't motivated by political grievances against U.S. policy. The U.S. removed a dictator that was keeping the lid on religious sectarian infighting. It erupted immediately into a civil war and that was further inflamed by third parties.

"The fall of Baghdad saw the outbreak of regional, sectarian violence throughout the country, as Iraqi tribes and cities began to fight each other over old grudges. The Iraqi cities of Al-Kut and Nasiriyah launched attacks on each other immediately following the fall of Baghdad to establish dominance in the new country, and the U.S.-led Coalition quickly found themselves embroiled in a potential civil war. U.S.-led Coalition forces ordered the cities to cease hostilities immediately, explaining that Baghdad would remain the capital of the new Iraqi government. Nasiriyah responded favorably and quickly backed down; however, Al-Kut placed snipers on the main roadways into town, with orders that invading forces were not to enter the city. After several minor skirmishes, the snipers were removed, but tensions and violence between regional, city, tribal, and familial groups continued."

https://en.wikipedia.org/wiki/2003_invasion_of_Iraq


> The U.S. didn't occupy a functional country and then destroy it.

But they did disband the Iraqi army and the whole Iraqi state!

>"The fall of Baghdad saw the outbreak of regional, sectarian violence throughout the country, as Iraqi tribes and cities began to fight each other over old grudges."

These are things the Iraqi Army and Police could have kept under control, had they existed. Instead, unemployed soldiers made up much of the sectarian fighting forces.

A country without a state devolving into anarchy shouldn't be that surprising. But to the US, it apparently was!


It wasn't surprising to the US. That was the whole point. To destroy the Iraqi oil industry.


Why would the US, the country dependent on oil prices being low, would want to destroy the oil industry of a major producer?


Depends on who you ask. There is a somewhat popular conspiracy theory out there [0] that claims Saddam (just like Gaddafi) got "removed" for selling oil in Euros instead of Dollars [1], which would have hurt the petrodollar in the long term.

[0] https://www.quora.com/If-Saddam-Hussein-and-Gaddafi-were-ous...

[1] http://edition.cnn.com/2000/WORLD/meast/10/30/iraq.un.euro.r...


Yes, that and what I said in another subthread.

Also, Sadam had outlived his utility.


First time I've ever heard this. I've heard lots of how the U.S. went in to steal Iraqi oil, but never about how the U.S. went in to destroy the Iraqi oil industry. Explain?


From what I've read, it was a deal with the Saudis. Iraq was reportedly planning, with Russian help, to modernize its oil industry. And to out-produce Saudi Arabia, and break OPEC. And to evade US sanctions by selling oil in other currencies.

There's also reporting about close ties between the Bush and Saudi families. Same private schools. Kids vacationed together.


They were also planning to stop using the $ in favour of the Euro to price their oil. US business interests did not like that one bit.


The U.S. didn't occupy a functional country and then destroy it. The outbreak of mass murder and tit-for-tat violence wasn't motivated by political grievances against U.S. policy. The U.S. removed a dictator that was keeping the lid on religious sectarian infighting.

The US got a lot of good will right after the invasion. It destroyed that by disbanding the Iraqi army, by failing achieve basic safety for the country and by managing the country in an incompetent fashion.

Moreover, the country was in horrific shape due to ten years of draconian sanctions. They were theoretically against Saddam but the average Iraqi suffered horribly - the looting of priceless Iraqi artifacts by destitute mob was an expression of what US sanctions had reduced the country to (along with Saddam himself, of course).


Revisionist if you don't have a clue what Special Forces/CIA et al do....


A dictator that the US had supported, initially because he killed lots of communists, and then fought a proxy war against Iran.


It's quite similar to what happened in Syria. When the Syrian government was torturing "terrorist suspects", for the CIA nonetheless [0], it was also considered a US ally.

[0] https://www.theguardian.com/commentisfree/2012/feb/19/syria-...


If the occupation force used cheap, throwaway drones then guerrilla tactics are rendered obsolete. But even that scenario would be inefficient.

The most efficient form of occupation is by getting the benefits of the occupation without the occupation: discretely hijacking a government + their media + most important companies, turning every citizen of a country into a debt slave.

Why fight millions of people if you can pay off a few dozen senators?


Such "asymmetric warfare" can be very effective: if defending costs much more than mounting an effective attack, then an underdog attacker can keep on bleeding the better-resourced defender or even eventually force the defender to give up.

The use of COTS drones might not become a trend though, since jamming/spoofing the GPS and the radio frequencies used would prevent the drones from being directed effectively. Customizing the drone radios would add to effort and costs, and require somewhat specialized skills.


I think there's a dangerous combination represented by the FPS Russia video with the drone-mounted gun. But instead of going for automatic fire, mount something like a TrackingPoint computerized rifle instead. This combination could be made even more dangerous by emphasizing low noise and observability, perhaps by using ducted propellers and fixed wings. If such a drone were inaudbile from 3000 feet, it could amount to an undetectable assassination machine, requiring rich and powerful people to have a 24/7 team of anti-drone specialists.

https://arstechnica.com/gadgets/2013/01/17000-linux-powered-...

EDIT: Before you accuse me of making up fantasy weapons, you'd best read the above link, and the one below. Computer aimed firearms that perform at "sniper in a box" levels actually do exist.

https://arstechnica.com/gadgets/2015/01/trackingpoint-shows-...


With a drone, what's the point being 1000m away ? You can put a few grams of explosive on a $30 nano-drone and detonate it right on the chest/head of your target. The “Slaughterbots”[1] video gives a good idea of what it could look like.

[1] https://www.youtube.com/watch?time_continue=191&v=9CO6M2HsoI...


Cause drones can be jammed/blinded up close. It's much harder to defend from a bullet fired from a kilometer away.


Jamming is easy enough if the drone requires realtime active control, especially if it's on well known frequencies and modulations - but autonomous flight using GPS, 9DOF accelerometers/gyros/magnetometers, and computer vision - is all current hobby-grade reality.

The main reason those are not (yet?) being used for terror attacks is that the skill level required to pull that off is quite high, and the people planning terror attacks seem to find it easier to recruit suicide bombers than geeks.

If any of _these_ guys: http://ardupilot.org/about/team - got "radicalized", I suspect they'd be very difficult to defend against...


Is it possible to jam the drones and not also end up jamming your own communications?


Yes. In its simplest form, radio jamming is just tramsmitting junk on the same frequency as the target. If you want to jam a couple kids using channel 10 on their FRS radios, all you need to do is put your own radio on that channel, ducttape down the push to talk button, and stick it in front of a laptop playing a 10 hour rick-roll video. Jamming RC aircraft, or GPS, or any other radio service is more or less the same in practice. The only difference is the type of transmitter you need to emit the jamming signal.

Radio jamming can be as broadband or narrow band as you desire. The narrower the band you target, the less power you need. The catch is, 1) you need to know what frequency you are trying to jam, and 2) this is not at all legal, so it would probably only fly in the middle of a warzone.


Realistically, a home built drone could be using anything for comms, including the civilian mobile networks or wifi band. Coupled with a tracking directional antenna, such a setup would be very difficult to detect or jam.


Action from distance could allow several drones set to fire on a target from different locations and calculate the projectiles to arrive at the same time.


Yes, and as I noted above, such systems are far less capable than something like a Hellfire missile fired from a Predator drone, so might be preferred to avoid escalation.


This is actually one of the more terrifying videos out there on this subject.


The FPS Russia video is a fake video / joke that you apparently failed to get.


The FPS Russia video is a fake video / joke that you apparently failed to get.

Your reading comprehension is poor. I'm only using the image as an icon/illustration to prime the discussion. The actual device I discuss is quite different.


Seems DW.com failed to spot that too...


[flagged]


Have you ever shot a gun in real life

Yes. In JROTC.

developed an electromechanical system more complicated than an RC car?

Before arduino was even a thing, I wrote a bit-banging controller for a converting tiltrotor VTOL. (We were using the GPIO of a "Gumstix" micro linux board to produce the PWM signals for something like 6 motors with partial automation to continuously mix the vertical-mode and horizontal modes of control, dependent on wing tilt, so bit-banging a microcontroller was the only way we could do it at the time.) The model tiltrotor project was headed by a friend whose father worked on the actual LTV XC-142, so it was quite complex. The complete tiltrotor never flew, but the "Gumstix" did fly on a delta wing plane.

Because what you just described is a fantasy.

More like fodder for near-future spy thriller Sci-fi, but it is well grounded in reality.

The TrackingPoint rifles actually did work, and there is one with an 1800 yard range. The controller on a Linux system is the thing which actually pulled the trigger, relying on the operator to hold the rifle in the correct orientation. When it calculated there would be a hit, the computer pulled the trigger.

https://arstechnica.com/gadgets/2015/01/trackingpoint-shows-...

Mounting such a device on a large enough drone should be well within the capability of folks like CIA contractors.


> The TrackingPoint rifles actually did work, and there is one with an 1800 yard range.

TrackingPoint is attaching a very sophisticated electronics package to an otherwise standard rifle. The 1800yd range is based on a .338 Lapua round. That round requires a pretty hefty platform to remain stable.

> The controller on a Linux system is the thing which actually pulled the trigger, relying on the operator to hold the rifle in the correct orientation. When it calculated there would be a hit, the computer pulled the trigger.

There are two problems with this on a drone:

First, I don't see any information (understandably) about the allowable limits of muzzle motion for TrackingPoint. I suspect the system is tuned to correct the relatively small oscillations the muzzle goes through in the hands of a human shooter. I would love to see the drone platform that could hold a .338 rifle that steady half a mile in the air.

Second, the system calculates the solution at the point of the trigger pull. There is a brief period between trigger pull and the round exiting the barrel where barrel movement can throw the round off. This is fine if the movement is predictable (as it would be for a trained human). I suspect a drone is going to struggle mightily with a .338's recoil.


Drones can shoot down, drones can shoot up, drones can operate in free fall as easily as powered, they can operate in perigee and apogee, butter side up and butter side down.

Look at the ways the OP is (charitably) correct, picking holes doesn't make it impossible.


The 1800yd range is based on a .338 Lapua round. That round requires a pretty hefty platform to remain stable.

Yes, I was thinking of something much larger than your typical consumer drone. I wasn't thinking of the little Mavic drone that lands in your hand. I suspect that it would be much smaller than a Predator, however.

I would love to see the drone platform that could hold a .338 rifle that steady half a mile in the air.

I should think there'd be some kind of gimbal involved.

Second, the system calculates the solution at the point of the trigger pull.

If you mean, "around the time the operator pulls the trigger."

This is fine if the movement is predictable (as it would be for a trained human). I suspect a drone is going to struggle mightily with a .338's recoil.

You're telling me that CIA contractors or other engineers couldn't engineer some kind of gimbal that would be more stable than a human operator? Making it lightweight and incorporating it into a drone would be difficult. I don't think it would be impossible or impractical.


Since we are all cherry picking details (or taking pot shots might be a better phrase)...

A TAC-338 sniper rifle weighs in at a whopping 11lbs. There are off the shelf camera gimbals that can easily hold a 120lbs camera dead still from as far away as you care to film.

I don't think stabilization would be an issue at all for a drone mounted rifle.


> I should think there'd be some kind of gimbal involved.

Random thought: If the drone can take a moment to set up and snipe from range, what about dropping some ballast down a telescoping pole? It may not help with yaw, but it'd reduce tipping.


Shoot straight down, instead.


> This is fine if the movement is predictable (as it would be for a trained human). I suspect a drone is going to struggle mightily with a .338's recoil.

I was dismissing the OP at first, but you convinced me it's viable. This one detail (that is really the worst I could find) is the kind of thing where neural nets excel.

One would need to fire a few shots on training, but now I think precise targeting (in there being a small enough motor step) is harder to solve than this.


It only requires that to remain stable if you’d like to quickly fire another shot. Otherwise you can just let the rifle move: the round had left the barrel long ago by the time it moves even a tiny bit.


We have recoilless rifles and Gyrojet rocket propelled bullets. It wouldn't take a motivated organization much effort to apply these to a drone to improve stability.


There are serious challenges to making Gyrojet style rockets as accurate as rifle bullets. These haven't been solved at manufacturing scales, so snipers still use rifles after all these years.

However, there have been demonstrations of a guided 0.50 caliber bullet.

https://www.darpa.mil/news-events/2014-07-10


Is there reason to believe that gyrojets could be made practical? I was under the impression that there were fundamental flaws in the concept.

A small-arms-caliber recoilless rifle could work really well on a drone, though.


Gyrojets never took off because of how expensive they are to produce compared to conventional ammo. ($10 vs $0.50) Their only advantages are gun size and recoil.

Micro drone guns might actually be the one use case for gyrojets, because recoil and barrel weight are the most serious issues.


I thought the major issue was that they were extremely vulnerable to being blown off course during the slow-moving launch phase. Basically, that even a mild crosswind would send it irretrievably askew.

I assumed that the price issue could in theory be solved by mass production economies of scale, if the other issues could be sorted out.


Yeah, aiming isn't easy. Dropping explosives is though.


True, but putting a bullet in one person is less likely to provoke a certain level of response. Drone bomb attacks are getting a bit too close to full warfare. A big enough drone-bomb is basically a cut-rate, very short-ranged cruise missile.


Easy and I believe already being done in Syria. I'll have to find video links but it's not new at this point.


Drone bombs became one of the favourite attacks by IS militants during the Battle for Mosul. They are a remarkably effective terror weapon, as unlike car bombs, they aren't stopped by walls or barriers. By all accounts they weren't actually particularly effective at actually causing damage.

They were basically dropping 40mm grenades (the kind used in under-barrel launchers like an M203) with fins attached.

Here's a video released by IS showing them purportedly blowing up a humvee with a grenade dropped from a drone: https://www.bbc.com/news/av/world-middle-east-39092883/so-ca...

Here's an article from the BBC with pictures of captured drones: https://www.bbc.com/news/technology-38663394


The drone with a machine gun is an ad for that years new CoD and is fake. That the article didn't acknowledge it shows they don't do their own research or are trying to mislead the reader.

https://www.themarysue.com/fake-cod-quadrotor/

Edit: spelling


But I don't think they claim that it was real?

"Whether reality or simply show, these examples make it clear that a good defense against drones is vital for both bodyguards and military in the field."


I find it funny that the "evil Russian" in the video is actually a gunsmith from Valdosta, Georgia.


This gun-drone is real, however, and may have already been deployed in Israel:

https://youtu.be/VaTW8uAo_6s

It carries a stripped-down AR-style carbine, and makes shots at what looks like 50-100 yards with sufficient accuracy.


I suspect a big drone could carry a one shot recoilless weapon I suspect with modern tech you could get the weight down to say > 10Kg

You can get drones that can carry over 40Lb an that's not far off a ww2 57MM M18's weight.


A number of things.

1) blocking GPS. 2) Electronic noise (mostly the same as blocking GPS, but needs lots more power) 3) thin nets. (bonus points for barrage balloons) 4) shotguns 5) flak (air burst anti-air shells 6) rain

Off the shelf quadcopters are fragile and not designed to cope with hostile flight. Loose a motor and your falling out of the sky.

Loose a GPS track and you'll either hover, or descend. This could be hacked to rely on dead reckoning, but that's inherently inaccurate. the sensors are designed to tell you what way up you are, how fast you are deviating from desired angle, they rarely have odometers.

A small > 250watt narrow band transmitter, Something with harmonics for 2.4gigs, or just a modified microwave magnetron with a decent directional antenna will destroy most hardware.

A thin fishing net is thin, light, cheap and almost invisible. The hardest part is keeping it in the correct place.


Before GPS, cruise missiles used photographs of the terrain they expected to navigate and a camera. The photos were state secrets back then; today they're easily accessible for the entire planet. Miniature computers are also far more power and more capable of the processing needed for navigating using photos and a camera. It'd take some R&D to develop, but it's certainly possible.

Also, "falling out of the sky" can still be an effective weapon if you plan for it.


While certainly possible, it would take the pool of people able to build it down at least an order of magnitude. I can say so with the experience of a guy who build drones and also tried working with some vision stuff for drone indoor positioning. It simply is far harder to do. GPS guidance you can do it in an slow evening.

Most of the measures intended to stop access to explosives, weapons and generally dangerous things are not able to do so 100% but simply make the cost of that access high enough MOST malevolent actors will simply not afford it.


The problem is that virtually all those things could harm, interfere with, or annoy human activity.

GPS is used for all sorts of legitimate human activity, so blocking it would cause problems.

Electronic noise could interfere with not just drones, but all sorts of electronic devices that humans use and need.

Nets, flack, and shotguns could certainly interfere with or harm humans, depending on where they are, and if there's some place they're not to let humans pass, then drones might be able to go through the same route.

Rain is annoying to humans and so plenty of places where humans are are protected against rain, so in those places this is an inadequate defense against drones -- though maybe sprinklers would work (but they can cause their own problems to useful electronics and might even cause an electrocution hazard to humans, depending on where they're used).


The problem is that virtually all those things could harm, interfere with, or annoy human activity.

In the context of how to respond and if possible disabled a weaponized remote controlled aircraft, I think being dead or seriously wounded also interferes with human activity. These proposals aren't-I didn't get the impression at least-meant to be implemented and operated persistently, but applied into purpose-built apparatuses to disrupt a hostile remote controlled aircraft.

That's a context I'm willing to trade having my phone work for a moment if there's a DJI Phantom dropping bombs over Boston (I beg humor for the hip-hop hyperbole) being taken out of commission before someone gets hurt.


The problem I see is that jamming GPS / wifi / mobile frequencies once you see a suspicious drone is too late. When the drone is already in view, it can find its target by inertial or camera navigation alone.

But jamming 1km around every possible target, every day as a precaution is going to be too disruptive. It may actually increase risk to life - by jamming 911 calls for any ordinary accident.


You are of course spot on. You have to asses the cost of implementing vs the risk/cost of the attack

For a spurious one time threat from an unknown actor, then most of these are overkill.

However against a known adversary with an MO for doing drone attacks, these might be legitimate.

An example, a sports event: stringing a fine net from the building/all weather lights, directional GPS blocking (its RF so can trivially be localised.) is cheap and proportional. "RF snipers" with 900 watt converted microwave "guns" (or more likely a beam forming radar type affair)


To the article's credit, none of these solutions address the "high up drone dropping a grenade on you in an area you can't secure beforehand" - simply because you won't know it's there until it's too late.


But that's not uniquely a drone problem. Aren't modern mortar shells wicked accurate?the trigonometry that governs them has been known for centuries.


i can't imagine how unguided shells can be that accurate, especially if you only have one shot. trigonometry helps, but there are way too many unknowns (mainly atmospheric cconditions) for it to work perfectly.


We have guided bullets today. A guided mortar is downright trivial in comparison.


agreed, reminds me of this xkcd :)

https://xkcd.com/1958/


Reminds me of the fallacy that Autonomous vehicles need to be absolutely infallible, when in fact they only need to have fewer fatalities per mile than humans.


Visual odometry and visual SLAM is absolutely mature enough for drone state estimation and control. It’s not off-the-shelf technology quite yet, but a skilled adversary could concoct something I’m sure. I imagine drone attacks today are likely going to be FPV, and then jamming of the video feed is possible again.


This is quite impressive:

https://www.youtube.com/watch?v=USYlt9t0lZY

https://arxiv.org/abs/1705.02550

And it's solving a somewhat harder problem - it's done with no maps apart from what it's generating in real time.


See Skydio As an example of a company doing this off the shelf already


An accelerometer and gyroscope may not make for a precise dead reckoning system, but if all you need is to get into a 1m blast radius, they might be good enough. Pair that with a smartphone with face/feature detection like those used for AR, and you might be able to navigate pretty much anywhere, no need for GPS.


IIRC exactly this was the basis of removing selective availability on GPS signals for civilian use. The original GPS spec had two bands / channels / something: one that was encrypted, for military use, and a "fuzzy" version for civilians that would vary ±100 meters in accuracy.

Once inertial guidance for missiles became more accurate than ±100 meters, there wasn't a point to reducing the civilian accuracy anymore, so they disabled SA in 2000 and went all-in on availability in 2007.

https://www.gps.gov/systems/gps/modernization/sa/


The V2 guidance system was an integrating accelerometer and a gyroscope. They could barely hit a target the size of London. The lack of accurate guidance made it a militarily ineffective weapon.


To be fair the V2 rockets never hit anything useful prinarily because the British had turned every german spy and were feeding them false information about the results of the attacks.


The rocket guidance system was not capable of hitting something more precise than "London", and of course the British tried to stop even that from happening.


Also interception by drone.

I know somebody who was having a nuisance/peeping tom issue with a neighbor and a drone. They had an unfortunate accident where their cheapo drone somehow collided with the neighbors.



Most of these can be countered by having hundreds if not thousands of cheap, fast and autonomous drones launching grenades from high or very low altitude. The swarm can loose a couple because of a fishing net or anti-air flak, but they can try another attack vector, loose a couple because of shotgun and try something else again.


Could a similar technique be used for defense? Program hundreds of drones to track and collide with enemy drones at high speed?


I don't know, you can build a racing drone for $200 https://www.youtube.com/watch?v=6MwKdHZ8OlM these go at 100mph and can change directions multiple times in one second.


It's worse, actually. There are motoless gliders that fly up to 545 mph, and then there are tiny jet planes that go up to 465 mph. You can jam the signal, but as soon as someone can put an autonomous vision system on them, it will become an extremely powerful weapon. You don't even need an explosive at these speeds, it can cut your target in half

https://youtube.com/watch?v=MoaWlKC3wIM


Frau Farbissina: We have developed the perfect assasination weapon! A silent glider capable of Mach 0.7!

Scott Evil: Great, now we just need to get the Venezualan President to give a speech at Torrey Pines or the backside of some other world class dynamic soaring hill...


Gliders can be dropped off balloons or airplanes. Their advantage is they are simple and cheap.


Yep - you won't get 500+mph out of that though. And the 500+mph dynamic soaring "gliders" are neither simple nor cheap.

(That's only relevant to this sub-thread's discussion, your point is completely valid in the context of the top level thread... Having said that - needing balloons or planes to get your home-brew autonomous weapons above your target is a whole extra level of logistics and opportunities to get observed/caught compared to launching consumer-grade drones.)


That's how the drone wars started.


If you know its coming. And if you have enough. What about 10 hardware-store drones with a few grams of explosive each, operating on dead-reckoning to reach their target? What about 20 of them? 100?


Id imagine a very directional antenna (or multiple ones to cover more bands and polarisation) with a 14 15 dbi plus a beefy wifi transmitter would be one possibility.

Just don't aim it like a rifle the side lobes might get nasty


A microwave magnetron and a satellite tv dish...

900-ish Watts of 2.4GHz will upset most off-the-shelf drones at quite some distance...


Nets would work to arrest and capture the drone, but to stop it you could probably get better results with a confetti like shot of something stringy and sticky. I bet long thin cotton or nylon thread would quickly tangle in the propeller, and sticky tape shards would disrupt the airflow properties of the propeller.


Honestly, I think underground, subterrainian cities are quite possibly the only silver bullet answer, as a correction for some of the emerging technology we'll be forced to confront, if things get out of control with everything laid out in the open.

There's this view, held by some, that with the atomic bomb, the cat's out of the bag, and the escalation of weaponry hit the ceiling in terms of organized conflict. But that there's also a non-linear side effect.

The outlook of this view holds that, if we indeed smacked into the ceiling, harnessing the same natural power that fuels the sun's daily dose of daylight, well, there's a whole vacuum of optional behavior between a floor of sharpened wooden sticks or hurled rocks, and a maliciously used hydrogen bomb, and all of human creativity to imagine interesting ways to negotiate that gradient between a single hostile neolithic ape, and nuclear age, mutually assured ICBM psychosis.

So maybe we got only a foot off the ground with primitive biological sabotage, or fumigated battle trenches dug into former farm fields, before we catapulted into the upper limit with the idea of multiple re-entry vehicles delivered in waves, to target thousands of cities simultaneously.

And what was the idea for how to cope with all your favorite cities being instantaneously incinerated?

That's right. Cave networks underneath mountain ranges. Going full dwarven on the problem.

So, the bonus side effect of going full AD&D dwarf character class, is that it also models well for self contained space stations.

Maybe closing the door to the biosphere, whilst certainly a tragedy of the commons it may be, possibly opens a window to outer space?


> And what was the idea for how to cope with all your favorite cities being instantaneously incinerated?

> That's right. Cave networks underneath mountain ranges. Going full dwarven on the problem.

I've played Dwarf Fortress, and I know how those games typically end. I think we have enough !!FUN!! on the surface.


Wouldn't a subterranean city by traversal with a map and sonar sensors? The only thing I could see a flying drone struggling with there would be the doors.



But drones can fly in your underground city, too...


The economy does. There are no off the shelf drone weapons. To develop them would require moderately skilled professionals. Moderately skill professionals in good economies do not engage in guerrilla weapon production because they have better ways of making money without the criminal liabilities.


Professional, singular. All it really takes is one person with an ideological axe to grind and a moderate amount of skill, and you have instructions that others can use to turn off-the-shelf components into weapons.

It's happened before. P.A. Luty wanted to make a point about British gun control laws, so he built a submachine gun out of ordinary hardware store parts and wrote a book ("Expedient Homemade Firearms") containing detailed instructions. He got jailed over making these weapons, of course, but the point stands.


And that's why Britain is overrun with submachine guns to this very day!


Mexico and Latin America are the main place these sorts of weapons show up, and there was a noticeable uptick in these types of firearms after the book was published. Most people aren't willing to flout the law and build submachine guns, but once the knowledge is out there it's a thing that outlaws can and will do.


Which is nice because then you get a super easy way to arrest them. It's much easier to arrest a gang member for having a weapon than catching them on the act. It is very common in Europe.


Moderately skilled teenagers with lots of free time can do plenty of things "just for the lulz". More skilled adults disgruntled by any reason (religion, politics, love, etc) can do some serious harm even when they could put those skills to far better use.


ISIS/hamas et al says otherwise.

War brings out terrifyingly brilliant innovations from the strangest of places.


Terrorism like this tends to happen in more impoverished areas of the world. And any innovation that is found in this pales in comparison to what the US military already has.


How is that relevant? The fact that IEDs wouldn't pass muster in a proper military doesn't mean we don't need tools to deal with them.

Having a better weapon doesn't stop a shitty weapon from killing you.


It's true.. when was the last time the US fought "proper" army regulars? Iraq in 2003/2004? Most war these days is asymmetrical.


"Having a better weapon doesn't stop a shitty weapon from killing you." True

I guess that im just making an argument about the modern west.


I don't think 1970-90s UK counts as "impoverished" the IRA/UDF did lots of innovation. Deadly destructive innovation.

Making bomb timers that lasted months/years to avoid detection was no mean feat in the early 80s


I'd agree that this is a fundamental factor enabling social peace (besides a culture of sincerity and trust), but aren't there in principle always incentives for competitors to eliminate one another? What if this becomes too easy to accomplish in a physical manner without leaving any trace?


What are the odds someone could pull off a drone assassination that leaves no physical evidence recoverable by the FBI? Surely some piece of debris will contain DNA, a serial number, some identifying information that would lead law enforcement to the perpetrators.


If the drone sets itself to fire it will be very hard to identify anything on it (DNA denaturates after 60 seconds at 200 °C). The problem about off–the-shelf components is exactly that they can rarely be traced back to a particular seller and could even be salvaged from stolen junk etc.


They don’t have to. We have crazy people who decide to shoot up schools and public areas with no concern with getting caught or killed.


It’s pretty trivial for someone with basic reading comprehension to find out how to make a self-igniting Molotov cocktail.


Just out of curiosity, had you already watched the embedded video ("Prototype Quadrotor with Machine Gun!") – of the quadrotor with the machine gun & the self-destruct ability – before you wrote your comment? The gentleman in the video didn't strike me as someone terribly concerned about product liability.



Ted Kaczynski. Timothy Mcveigh. Dzhokhar Tsarnaev. Mark Conditt. Etc


In the end there's really no good way to electronically block against a drone. Someone could preposition some infrared lasers pointed along the expected flight path and the drone could use those to perfectly navigate to the target. Or a single laser that tracked the drone by scanning the sky and the drone acknowledging when the laser was pointed at it. Or I'm sure a dozen other schemes. Someone willing to spend a year tinkering could come up with a drone that is only stoppable by directly disabling it.


Many will not like this, but: off-the-shelf shotguns in the hands of competent shooters, among other things.


Check: Tom the dancing bug - Goodguy with a Gun


Actually, simple off-the-shelf weapons are extremely effective against drones. A 12ga loaded with #9 pellets can easily defeat any off-the-shelf drone. And to boot, the pellets fall harmlessly back to the Earth if fired into the air.


but can it defeat a dozen of them at once? a hundred?

the big shift that drones represent is replacing trained soldiers with direct industrial output. the future is swarms.


"off the shelf" vs "off the shelf"


The fact that most people value their freedom and are not looking to commit crimes. There are many many things that could be done, but if done mean essentially destroying your life. Most people don't want to destroy their lives.


?

Most people don't want to destroy their lives by shooting someone with a gun, yet it happens every day.


Okay I guess then we should do something about all them guns. Point there being that guns are readily available and this seems to be maybe slightly more dangerous in certain cases.


My point was not about guns in particular. People use knifes for that too, or rocks, or baseball bats, and we aren't going to ban those.

My point was that there are people who want to murder, and drones will be very suitable for that, and the fact that 99.99% of people don't won't really matter.


One griefer with a website that offers exploding drone swarms to a given target? 'Most people' isn't everybody, and it only takes one. And the website could be anonymous (bitcoin anyone?)


It's a problem of scale. Say 0.0001% (number pulled from my rear pocket for illustration purposes) of the population want to engage in 'recreational terrorism' or 'orc-work'. If you have a population in the hundreds of thousands or low millions, there just aren't enough of them to have any direct impact on the big picture. But once you get the population high enough, once the absolute number of evil idiots is high enough to cause real systemic damage, even though they are a tiny percentage of the whole.


One tiny group of people hijacked 3 planes, and we're suffering the fallout decades later.


I am almost certain we have laws on the books to deal with that. The point is identifying that before they do anything.


Deal with what? A website that relaunches in the cloud from a different bitcoin account every 20 minutes? Anonymity is great from a personal-freedom point of view, but not so much when assassins start advertising their services out in the open.


I make a kidding-not-kidding joke about skill with bolas making a resurgence as a counter to drone harassment.


Something like this: https://en.m.wikipedia.org/wiki/Krasukha_(electronic_warfare...

and something like this: https://en.m.wikipedia.org/wiki/Pantsir_missile_system

(Not familiar with the western equivalents but the idea is the same.)

A small drone does not have the ability to carry the shielding required to protect itself from interference, which can include jamming GPS and even taking over controls of the drone to capture it.

A larger drone is easy target for AA systems.


Those failed to protect Russian base in Syria even against low-quality homemade drones.


They have not failed - not sure where you get your information from. Maybe you are confusing with the mortar attack earlier this year?

From earlier this year with pictures: https://www.independent.co.uk/news/world/middle-east/russia-...

There were several attacks repelled in the last month also but I only have Russian language sources (maybe South Front will have some more details in English - coverage of Syria in Western media has been pretty sparse in the last months with all the government successes in Daraa).


I would imagine that defensive sensor drones would be the first priority. You really need to detect enemy drones and having a swarm of sensor drones able to follow potential targets would be really good. Radar and microphones are going to be interesting when applied to drone detection.

Once the target is detected, a rather large cheap swarm deployed on the attack vector to block projectiles and intercept the enemy drone would probably make the most sense. The drones can absorb the hits and blow up when they reach their targets.

ECM-based solutions do seem to have problems with also taking out the good guys equipment and capabilities.


The answers I read here are classic over engineering. Sometimes the simplest answer is the best answer, and you don't need to build a complex system. That is to say the same methods that are used to defend against a gun, either another gun to disarm the assailant (be it a person or drone) body armour. Killing the GPS, etc. are all good methods too, don't get me wrong, it's just that they may be more Factory.getFactory().getFactory().getInstance()... type of solutions. Initially just the same methods should work.


Anything that blocks radio control or GPS.

The same things that defend against off-the-shelf radio airplane weapons - a risk that has spanned decades. A couple of dozen dollars and some foam core creates an equivalent risk (or greater, since it has a longer range due to being more efficient than multirotors).


> Anything that blocks radio control or GPS.

Don't they have inertial navigation if they can't get a GPS signal, and pre-programmed routes that don't need radio control.


They give you what direction is up, but not really where you are and how fast you are going.

You need odometry for that. There are opticalflow modules that will allow you to do that, but they cost power.


Power draw by a SoC like Nvidia tx2 (can probably do it with a cheaper chip too) for camera based navigation without GPS is not a big issue, the motors are far more power hungry


Practically speaking though, a mortar would be cheaper, simpler, and more deadly. Same(ish) range, high rate of fire, designed to be anti-personnel, provides plenty of time to vacate the area after initiating a bombardment.

Bringing tech into the equation doesn't add much when you're talking about a payload in the single-digit pounds region.


Inertial navigation is possible, but the sensors are fairly low quality for anything over a couple of dozens of feet.


Can it be combined with scene reconstruction from cameras? My simple iPhone seems to extremely good at it in the AR kit.


At the cost of money and payload loss (or a range reduction), absolutely. If you're willing to spend engineering hours and thousands of dollars per drone, you can absolutely have nearly fully autonomous drones that don't rely on GPS.

Still, not really a new risk, however.


Ok, then, laser turrets.


Other drones with lasers.


Actual security by design. Hold events with high-value targets //indoors// with no open roofed areas near by.

This might mean throwing a net over the inner alcove or erecting a temporary tent to block the outside facing windows.


A tactical nuke that emits a strong EMP.


Underkill


Since nobody's mentioned them yet: Net guns.

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


NULL pointers


and after this decade:

    undefined is not a function


Not much those in Rust


I imagine a scaled-down version of the U.S. Army’s C-RAM, or Israel’s Trophy system, could work quite well for VIPs. Some drones can be VERY fast (think hobby-sized jet engine fast), so you’d need active radar and full automation to stand any kind of a chance.

Perhaps you’d use shotgun shells loaded with rock salt, so any overshoot into a crowd would be less-than-lethal.


The answer is obvious: anti-drone drones. Military units and major political figures will need to have swarms of anti-drone drones patrolling their vicinity.

Things like auto-targeting lasers might work too, but the way the way to fight technology is with better technology.


When we are talking about one-off drone attacks, sure there are counter-measures. if we are talking about WW3 and China's capability to produce millions of autonomous, AI driven drones per day? No, there is no effective counter measure.


Other drones, obviously.

Whoever has more money to purchase and deploy drones wins. Same as it's always been.


Shotgun with birdshot.


Serious question: why is it hard to make small, anti-drone missiles? I assume it needs to only be within X distance of target and detonate to disable a drone.


Automated missiles at every public building, gathering, and billionaire's motorcade? What could possibly go wrong?


If you haven't read Daniel Suarez's science fiction novel "Kill Decision", now is the time


Directed energy with automated tracking.


Also probably the same machines that are installed for crowd control already a lot of military installations (microwave emitters) can do this already.


Explosives are heavy. Are drones a realistic delivery vehicle?


Not for anti-tank, but they have enough power to lift a hand grenade and throw it onto personnel or if very lucky into open hatch of a tank [1].

[1] - https://www.reddit.com/r/CombatFootage/search?q=drone%20gren...


The US military's current hand grenade - the M67 - weighs 400 grams, has a kill radius of 5 meters, and a casualty radius of 15 meters. The DJI Inspire 2, chosen at random, can carry 790 grams.

Yes. It doesn't take that much explosive to kill someone.


It can likely carry nearly 3-4 times that but you’ll burn out your engines, but on a short one way trip that is not a concern.

A claymore mine can be fitted to the bottom of most drones it’s much more effective and it’s electrically triggered.


More off the shelf drones? Seriously.


concealing your location is probably the best defense


Nothing.


[flagged]


I think you mean the German V1 :-)


Hmmm V1 had no remote pilot, it doesn't count :-) Japan's kamikaze are closer to the real thing (it's not a remote pilot neither, but at least they came close to the enemy whatever its defense, much like a drone) :-)


the US had crewless bombers piloted via television for suicide runs in germany:

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

they were ineffective, though

regarding your original comment: the topic at hand is weaponized commercial drones, things that can be assembled by semi-skilled enthusiasts. predators cost millions of dollars per and use satellite coms and guided munitions. isis uses off the shelf quadcopters with mortar shells. the latter is essentially the 3d printed gun of assassination tools -- there may have been forebears, but it's the accessibility and ease of deployment that is important and alarming.


It was autonomously controlled to run on a set path and to dive when I thought it had reached the target.

drones don't have to have a human controller.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: