kimbosbread

As was discussed a few weeks ago, welding a rail onto a mortar, and attaching a combo gps/gyroscope to get absolute position and orientation would be relatively simple and allow mortars to have much better precision and faster targeting loop when combined with spotter drones. That said, I’m really liking the idea of GLSDB, but smaller and launched using a mortar charge.

Much faster, in theory. Frame is one piece, can be injection molded plastic or carbon fiber at large scale. Motors just drop in, same with compute module and cameras and warhead. Balancing it is a thing. Assuming sufficient supply of these components, an assembly line of a few people should be able to crank out a hundred per day. Transmitter range from base station (send control signals) , and transmitter range on drone (send video and diagnostics back). So if you have mast at the base station, you can get a few more km, and if you have a repeater drone, some more. But you figure these drones have 20-40m endurance at 60-100kmh speed, so there’s more possible range than is capable with a small, low transmitter. Autonomy is how you solve it. My personal opinion is repeater drones are probably the best option as they protect the operators better.

Most modern technology (computers, factories and otherwise) is beyond most people. I would submit that basic algebra, physics, etc. are also beyond 90% of the population. A decent amount of the population could figure how to bang some rocks together, if it came down to that.

Paging Iain Banks to the blue telephone… But seriously, I think this is a route you could see the US taking. The quality of politicians is atrocious, especially at the local level for many reasons, and this might just be a way out of it.

DARPA was working on this kind of system at least 5 years ago for drone swarms, and obviously this was used for the famous grand challenge that was around 20 years ago. You know how combat video games look pretty real? This is the same thing, except the computer is playing the game. Seriously check out https://microsoft.github.io/AirSim/ for something that is stone age by today’s standards. I honestly think assembling the training data is the bigger challenge, and I think LLMs offer a great opportunity here, where you can in effect describe a scenario or battlefield with words, and have it generate a few hundred and then humans can inspect and make sure it’s not completely unrealistic. How does AI work? We are talking about mostly computer vision here, so this is pattern matching plus a priority list of targets, and maybe a playbook of multi-drone attacks. This isn’t an LLM hallucinating whatever it hallucinates. You can run the same kind of tests other missiles get, and you can look at what FPV operators in Ukraine are doing. You can even use human FPV pilots as your baseline, and train your models based on what they do. Look, the evolutionary pipeline is pretty obvious: FPV drone, with pilot and transmitter in a hide or trench Same setup, but the transmitter/base station can now identify targets and let the pilot select them, and then the drone does the rest (but controlled by the base station) Push some of the computer vision hardware onto the drone, so it doesn’t need to have an always stable link to pilot ??? Autonomous death swarm!

Very fast. It is being trained in virtual (Look up Nvidia Omniverse and Microsoft’s drone simulator successor to Airsim) and real environments. If you can basically train 100 years of engagements in a few months, and supplement that with real training exercises, you are at looking at a quicker procurement cycle than pretty much anything else the military can muster. Fielding it is simply downloading the software and making sure the singature is ok.

Wrong. A smart grenade can fly. We’re pretty close to having these proliferate. I dunno, given the proliferation of drones they might rather defend windows first. I think this is wrong too. The revolution that allowed small quadcopters (way more computing power on a small, energy-efficient chip) also allows lot of small, cheap sensors. If you can just bombard a building with hundreds of sensors, you can find out where soldiers are. Sniff for CO2, look at termal signatures, measure vibration (footsteps, heartbeats, water in pipes) and use your cheapo compute for sensor fusion. Small area seems like a death trap.

Re corruption in Ukraine, someone involved in the post 2014 - pre 2022 training etc told me that around 80% of the US aid just went poof.

That’s all good and well, but ignores the medical, educational and social realities on the ground. The average reading level in the US is 6th grade or so, and public school is significantly worse than it was. It’s not really possible to fail out now. The fattest kids when I was in grade school say 30 years ago would now be the skinniest kid (recall Friends where fat Monica had to be made progressively fatter to size up with reality). And the ADHD med use is kind of nuts. There may be a silver lining in the ability to play on a smartphone which might crossover to drones, but most kids now have less computer sense than those 20 years ago who were in the sweet spot for learning about computers as you were still exposed to a file system, had to learn how to pirate software, install linux etc. But, a motivated kid can learn more easily now with all the info online. Now for the UK, I recall a decade ago Sihks were excited about having their own regiment. Seems like they could get plenty of motivated, capable young men and women for that kind of regiment.

I have heard the woke-ism is destroying the US army trope, and while there is some truth to it (lots of conservative white guys in fighting units, lots of senior officers just tired of politics), the bigger problems are that we have had 20 years of pointless wars fighting sand people (not a worthy opponent), and the modern youth are fatter, druggier and less mobikalizeable than before. China has the other side of the coin- spoiled single boys, and families where there is a single grandchild. Not exactly great to get that kid’s face blown off by a drone. This is the biggest upside to Replicator and the whole drone-ize everything in my mind. The recruiting problem isn’t a problem if we (or the enemy) don’t need as many people, and we can scale training and learning as fast as we can scale our datacenters. Obviously that’s a bit sci-fi, but for some of our problems we may need sci-fi solutions.

Has to be a combination of several factors: Russia’s weapons aren’t that precise They don’t have the ISR assets available for finding large numbers of targets and confirming they are hit (how many optical spy satellites do they have? 2?) A lot of Russia’s behavior is performative and top-down, that is, send a bunch of mobiks over the hill, or fire your quota of artillery rounds. If Puting orders a big strike, just shoot all your missiles. Ukraine won’t publicize any success Sure, Russia has certainly destroyed some strategic targets (Hrim components, training camps, power plants), but simply doesn’t have the precision and quantity and ISR to really knock out for example all power stations in Ukraine. They tried last year, and they couldn’t do it. EDIT: They’ve lost tons of assets on the ground, thanks to SBU, using SF as regular infantry, losing most of their pre-war professional military. That has to play a role too.

You say that, but the Black Sea will have some epic wreck diving after the war. I know divers who are frothing at the loins at the thought of diving the Moscow (which is quite shallow).

Is the military actually more effective (despite losing element of surprise, territory, best soldiers, formations and equipment), or is this merely a manifestation of Surovkin understanding primacy of denial, and consequently mass not working?

A good saboteur might. But I’m the kind of person that gives money to the worst street musician on the block, or the most annoying pan handler in a tourist area. Otherwise known as “Somebody else’s problem”. Which is great, until you notice a new moon overhead that wasn’t previously there!

One theory goes that UA attrition of RU air defenses over the last 6 months has forced RU to move more expensive assets (AWACS) closer to the front to help with radar coverage.

That’s the frustrating thing about a lot of the tech that we have discussed: You don’t even need a lab; between a buddy and me we have enough electronics test equipment and spare ics and breadboards to put a demo together. I’m curious about using one of the 9dof sensors (used in VR goggles, watches, phones) and how you’d zero it for whatever weapon you attach it to. It really does seem like a long weekend project to get an MVP working.

A friend of mine (a furry) dragged me to a fantastic party in SOMA (San Francisco) where I ran into a trans coworker of mine (also a furry), who was extremely drunk and remarked how she loved our employer, but one small improvement would be urinals in the womens’ bathrooms.

Like many of us have said, look at the logistics and training footprint! Plus no tinnitus and TBI from firing artillery, no destroyed joints from lugging around 40kg shells. Plus no great counter-battery strategy.

We have a common rail system, thankfully! Protocol is un-necessary for MVP: Just have the drone guy tell you the GPS coordinates, and plug those into your universal firing computer. Drone-controlled firing and bracketing is version 2. I like the idea of the cell-phone that attaches to the rail. User interface is easy: Enter GPS coordinates of weapons system, or use onboard GPS Enter type of weapons system Enter target GPS coordinates Aim weapons system, and the phone display shows you crosshair representing aim point of weapons system, and then another crosshair representing aim point to hit target. Line them up, phone makes a sound or flashes and you pull the trigger. EDIT: So here’s the challenge: How do I zero our cell-phone firing computer? Do I need to have a known zero? Can I just estimate things based on accelerometers and a laser level and say “meh good enough” and hit stuff 5km away? EDIT2: Is this even worth the effort of developing and building, when I could just have more kamikaze drones?

100% I’ll go even further and say this should exist for all weapons, be it hand grenades and rifles on up. The drone-enabled ISR should be able to tell you “throw a few grenades over this wall”. Once we have heads up displays (or basic bitch huds in the rifle optics) there is not reason you can’t project an aiming point that gives you a nice ballistic arc. That’s the minimum viable product version of this, and it should be almost trivially easy to write software to compute this on the drone side that calculates rough GPS coordinates, and can bracket fires. The challenge is the aiming interface on the ground-based weapon honestly. Even if it’s manually aimed, how do you show the soldier where to aim it? AGLs and m2s aren’t really set up to hit a certain GPS coordinate, so obviously a computer controlled turret is great, but that’s an extra level of computation. Maybe you strap an ipad on top of the thing and that gives you an augmented reality aiming point? Easy to write the software for too, vs some crazy proprietary hud.

Also I’m liking the 120mm mortar as a replacement for tank direct fire more and more. It should be possible to deploy a loitering munition from one, or a super ****ty GLSDB analog, but mini and direct it from the overwatch drone.

This is where our UGV would shine. Give it some 1 shot 40mm grenade tubes and a machine gun and have it go to town. That said, I wonder if a better weapon wouldn’t be to just douse the position with gasoline or kerosene from the UGV. Don’t even light it on fire, just let it evaporate and freeze everybody in the position to death extra fast.

As much as a I like old man’s war remixed with ender’s game, the problem as always is the data link. If we had a magical laser or satellite link like real countries use to control drones, sure, no problem. But for small el-cheapo drones, that just won’t exist without a robust repeater network, which could absolutely be built. The real trick would be to allow pilots to fly from other countries, assuming latency is low enough. I’d love to go blow some Russian soldier face off to work out my aggression before morning meetings. I imagine people would happily pay for the opportunity to do this, in fact./

The bottleneck is bandwidth. There are a finite number of control (1GHz) and video channels (2.4/5.8GHz, plus analouge video channels), and this is exacerbated by EW. So you cannot control that many drones, over that large a distance, unless you have a lot/fancy antennas, and all the drones set up for it.

Taiwan has a very strategic location, so fabs aside there is lots of value for China. Whether it’s worth fighting a war over, well, that’s Winnie the Flu’s decision to make. In terms of workforce, Taiwan has very low pay. In the past, China got a bunch of Taiwanese engineers because pay in China in the high-tech sector was quite good, especially for people they want to poach.