3 ms·
Drones are effective, but nothing is ever cheap on the battlefield. While the warhead that ultimately kills the target is often cheap, the real cost comes from
by jltsiren 4y ago
Drones are effective, but nothing is ever cheap on the battlefield. While the warhead that ultimately kills the target is often cheap, the real cost comes from the system that picks the right target and delivers the warhead there at the right time.
Cheap commercial drones have a short range. In order to use them effectively on the battlefield, each drone needs an operator, who must also be on the battlefield. Those operators then need other people for watching their backs, coordinating their actions, and handling the logistics. Those 1000 cheap simultaneously launched drones are now an entire brigade on the battlefield, which is definitely not cheap.
- ncmncm 4y agoThe most effective use of drones so far in Ukraine has been to direct artillery fire, converting what was an area weapon to a precisely targeted weapon. Just a couple of days ago, a single artillery brigade eliminated an entire tank battalion at a river crossing. Generally, if you can wipe out the fuel trucks, they have to abandon the tanks. An ordinary grenade suffices to blow up a fuel truck.
- denton-scratch 4y agoI envisage a swarm of say 50 cheap-ish drones with a "swarm leader", intercommunicating using some kind of short-range mesh radio. They'd be disposable, single-use. They could be launched by a squad of say, 10 men. The swarm leader could be human-piloted or autonomous. Autonomy would be a matter of software and sensors; once you've invented it, it would be dirt-cheap to implement. A swarm of 50 would be enough to baffle air-defence systems capable of tracking multiple targets. I have no idea if this exists, but it seems an obvious idea; if it doesn't exist, it must have been tried and found wanting.
- KickKat 4y agoThis guy made an autonomous drone swarm of 12 drones similar to what you describe. The overral objective was for any drone within the swarm to immediately chase and ram into the first human it could locate. The processes were automated and could be triggered while unattended. For reasons the video explains, letting 12 drones all try to independantly do the same thing at the same time was not going to work, so he came up with ghis structure. 1 server that processed the video feeds from the drones using AI trained to recognize humans/human faces 1 C&C computer that sent master commands and stuffs .... and maybe the computer flight code ...? 12 drones assigned to their own LAN making them a virtual group 4 of these drones were designated as flight leaders. They sent data from their cameras to the video processing server and recieved targetting results back. Sent target and flight instructions to the two drones assigned to them. 2 drones are assigned to each flight leader drone forming a single unit. They only communicate with their flight leader. They do what it does. I know ... it isnt anywhere near what would be required for combat use, but come on... in the last test, his drones did what they were supposed to do; every single one of them automatically attacked the first human being they detected immediately. Better yet, they did it in unison. More better of all, all of them him. This guy did this with cheap, off the shelf parts, by himself, during his downtime at home as a joke for Youtube. https://youtu.be/Hu3p5ZR_i5s https://youtu.be/Hu3p5ZR_i5s
- imtringued 4y ago>during his downtime at home as a joke for Youtube. I'm pretty sure this is how we are going to get sophisticated autonomous weapons. Everyone is screaming and shouting how dangerous they are and then some dude wants to get a quick laugh or prove everyone else wrong and inadvertently creates a high quality weapon.