Live data from Hacker News

Drones Raining from the Sky in Zhengzhou

twitter.com

81–90 of 131 posts

Re: Drones Raining from the Sky in Zhengzhou

#81
post #51

Earlier quoted context omitted.

I am neither American nor a gun nerd, but from what I "overheard" on the internet a single Phalanx system should be able to take out drone swarms at incredible rates?

https://news.ycombinator.com/item?id=27440924 >>The Block 1A and newer (pneumatic driven) CIWS mounts fire at a rate of 4,500 rounds per minute with a 1,550-round magazine. >>Shells fired by the Phalanx cost around $30 each and the gun typically fires 100 or more when engaging a target.[14] >So it's about three grand to zap one drone, and you get 15 shootdowns before a human needs to shovel more ammo into the turret.…

I think for drones of this kind you can set up mini phalanxes. They don’t quite move at the speed that a missile moves, and the caliber necessary to destroy a drone is much lower than that necessary to disable a missile.

Re: Drones Raining from the Sky in Zhengzhou

#83
post #76
post #60

Earlier quoted context omitted.

Actually proximity fuzed rounds should work the best, each taking outv many swarm memvers. Its not like these things could be armored rather than as light as possible.

Proximity fused rounds do have the significant downside that using them near built up areas will cause significant collateral damage. For example, if a drone swarm attacks an election rally somewhere in the US then you can't really detonate airburst shells above the crowd to take out the drones, since you would do more harm than good with that.

Would you? What if the drones are carrying grenades?

I mean, I can see how it’s less than ideal, but I don’t think it’s safe to assume you’d do more harm than good.

Re: Drones Raining from the Sky in Zhengzhou

#84
post #45

For those who like drone dystopia nightmare fuel, this video made me think of this disturbing modern weapon: https://en.wikipedia.org/wiki/STM_Kargu :/

What defence could, say, a traditional military base, mount against a swarm of, say, 1000 of these?

I would think "old school" air burst flak could be enough... you only need a thick enough cloud of shrapnel flying around.

Re: Drones Raining from the Sky in Zhengzhou

#85
post #71
post #65

Earlier quoted context omitted.

Maybe. How much would you bet against open source projects like Ardupilot having their code swiped and used by defence contractors or weapons manufacturers. What are the GPL requirements for making the source available to people targeted by drones running stolen GPL code?

Do you think the Ardupilot people were blind to the military applications of what they were making? I personally can't imagine anyone bright enough to work on drone autopilots to be simultaneously ignorant enough not to realize that a drone could be used for harmful applications.

I know several people who've contributed to Ardupilot who are 100% "hobby guys". I've modified Ardupilot (and BetaFlight and iNav) for my own hobby use (but nothing really worth or in a state to contribute back).

While it's true drone technology "could be used for harmful applications", that's also true for, say, compilers. Do you hold the gcc/clang developers to account for being "blind to the military applications of what they were making"? Or OpenStreetMap contributors? Or the engineers who designed the HiLux? Or the Casio F-91W watch designers? None of those were "designed for military application" yet all ended up being used for them.

Where does the line need to get drawn? between my rights to design/build/fly drones as a hobby, commercial/industrial rights to use drones for photography or cinematograpy or crop monitoring or search and rescue or shark surveillance at swimming beachs - and the need to restrict "military applications" of that?

Are people who, say, write blog posts explaining how to use microcontroller RTOS libraries for hard real time guarantees, with mentioned examples like nuclear power plant control "blind to the military applications of what they're making"? (Yeah, sorry. I profile stalked you a little there...)

Re: Drones Raining from the Sky in Zhengzhou

#86
post #5

It's hard to tell from the video but arent they supposed to fail gracefully in a controlled descent and not drop like rocks?

It seems like most are doing that but some are dropping rapidly, although not in a freefall.

One of these drones in the second video is pretty much upside down as far as I can see.

Re: Drones Raining from the Sky in Zhengzhou

#87
post #71
post #65

Earlier quoted context omitted.

Maybe. How much would you bet against open source projects like Ardupilot having their code swiped and used by defence contractors or weapons manufacturers. What are the GPL requirements for making the source available to people targeted by drones running stolen GPL code?

Do you think the Ardupilot people were blind to the military applications of what they were making? I personally can't imagine anyone bright enough to work on drone autopilots to be simultaneously ignorant enough not to realize that a drone could be used for harmful applications.

GPS is military technology designed that for that.

Re: Drones Raining from the Sky in Zhengzhou

#88
post #83
post #76

Earlier quoted context omitted.

Proximity fused rounds do have the significant downside that using them near built up areas will cause significant collateral damage. For example, if a drone swarm attacks an election rally somewhere in the US then you can't really detonate airburst shells above the crowd to take out the drones, since you would do more harm than good with that.

Would you? What if the drones are carrying grenades? I mean, I can see how it’s less than ideal, but I don’t think it’s safe to assume you’d do more harm than good.

If you can say for certain beforehand that the drones are carrying grenades, then _maybe_. But if they are "just" carrying a sharpened stick to try and skewer whichever politician is holding the rally with, then killing a large amount of the audience will be really unpopular Also, not killing them but allowing the politician to be killed will be really unpopular too. Lasers on the other hand don't run out of ammo, cause much less collateral damage and are quite effective against lightweight drones.

(Of course by the time you have to guard political rallies so much that "grenade attacks by drone swarms" becomes a reasonable threat scenario, you may have to rethink things and probably just cancel the rally for now.)

Re: Drones Raining from the Sky in Zhengzhou

#89
post #85
post #71

Earlier quoted context omitted.

Do you think the Ardupilot people were blind to the military applications of what they were making? I personally can't imagine anyone bright enough to work on drone autopilots to be simultaneously ignorant enough not to realize that a drone could be used for harmful applications.

I know several people who've contributed to Ardupilot who are 100% "hobby guys". I've modified Ardupilot (and BetaFlight and iNav) for my own hobby use (but nothing really worth or in a state to contribute back). While it's true drone technology "could be used for harmful applications", that's also true for, say, compilers. Do you hold the gcc/clang developers to account for being "blind to the military applications…

This isn't really a new thing - the military can and will use open source code to more effectively kill people. That happens with every new technology. Drone tech is going to be the biggest killer this century assuming that we don't have a war where ICBMs get deployed.

Zyklon B was a pesticide, and all that.

Re: Drones Raining from the Sky in Zhengzhou

#90
post #85
post #71

Earlier quoted context omitted.

Do you think the Ardupilot people were blind to the military applications of what they were making? I personally can't imagine anyone bright enough to work on drone autopilots to be simultaneously ignorant enough not to realize that a drone could be used for harmful applications.

I know several people who've contributed to Ardupilot who are 100% "hobby guys". I've modified Ardupilot (and BetaFlight and iNav) for my own hobby use (but nothing really worth or in a state to contribute back). While it's true drone technology "could be used for harmful applications", that's also true for, say, compilers. Do you hold the gcc/clang developers to account for being "blind to the military applications…

I don't mind the profile stalking, but if you stalked me more you'd have seen that I'm a former weapons engineer for the Dutch navy, so making things with potential military applications does not bother me as much as you might think. (Ironically, that arduino CoPilot library is based on software specifically designed for drone applications, so does that count as military software redirected towards civilian applications?)

In any case, this discussion is pretty similar to the "AWS is running my FOSS database as a service without paying me" debate. If you don't want your software to be used for certain applications, don't release it under a license stating that you can do anything with it. If you think the good that you do for the world by releasing some good new tech is greater than the harm done to the world, spread it as widely as you can. Ardupilot is in this category IMO: even if some militias try to make their own combat drones with it, the overall good done by making drone technology available to all could well be greater. The idea that you can make something "perfect" (that is with no imaginable harmful applications whatsoever) is probably not viable in this imperfect world filled with less than perfect people. Even very harmless things like teddy bears can be used to enable violent acts, but that does not mean we should stop making them.

Post reply on HN