Tux

Ok, well I will have to take the hit on not being fully understood. I will think about it some more. I’m not entirely sure the idea I’m trying to communicate is valuable enough to justify the thread-space it’s consuming, at this point.

I appreciate the challenge but their utility doesn't have to be in question any more than any other weapon system. It's their deterministic necessity that undeniably will, one day, be questioned. Not all unmanned systems - I am not saying we try to "go back to the way things were" - but the most egregiously dangerous and offensive ones. My very first point was a reflection of the fact that C-UAS will absolutely be highly competitive, dynamic, and ever changing. So I mentioned that it might be an idea for forward-thinking nations to focus on autonomous C-UAS now, even more than on autonomous ground-attack drones. Get ahead of the game. Establish and then try to maintain C-UAS superiority. Lead everyone to question the deterministic necessity of these things as soon as possible. Once you do that the imperative to develop and use them is weakened and maybe people will fear the systems more than they will the consequences of not having them. Because, let's be honest, fear is always the arbiter of this kind of thing. I understand that this has been and still occasionally is challenged, even on this thread. It is not the target of my argument, though. Please, yes, let's navigate this new reality. As a starter for 10, how about we give absolute priority to autonomous C-UAS?

I totally agree and am focusing on nightmare weapons which hunt down individual human beings in order to kill or maim them. Autonomy against vehicles, factories, equipment and other unmanned systems will be ubiquitous and I think relatively uncontroversial. It's when these things start being used specifically to kill people that the world will cry abomination, just as they have done in the past with other systems that lead to outsized (even if unintended) risk to civilians (cluster munitions, AP mines, etc.) or which offend one too many animalistic sensibilities when they are used (flamethrowers). I understand that none of those weapons were considered deterministic at the time but it's the reaction they induce which I think they will share with human-targeting drones. So, once such drones can no longer be considered deterministic, the momentum could gather to outlaw autonomous targeting of human beings. All very hopeful, for sure. I am just raising the possibility in light of the fact that the most offensive type of UAS won't be deterministic (and therefore necessary) forever. I agree but don't think it will impact the weapons that people try to regulate, once they are not really necessary. Many of the attacks you are referring to are already considered warcrimes, after all... Granted. No argument from me against autonomous attacks on infrastructure, etc. My suggestion is that semi-autonomy *could* be reserved for targeting humans in the long term and once C-UAS has become deterministic to the point that whoever wins the C-UAS fight can maintain that dominance and effectively choose whether they use fully-autonomous-hellborn-head-poppers or not. This is similar to paxromana's point. Someone, somewhere, will try to autonomously attack men, women and children who wear the wrong type of clothing or use the wrong vowel sounds. I get it. However, if dominant and widespread C-UAS systems exist, then that needn't be a weapon of choice for whichever corner of humanity ethics end up sheltering in. That's all. My prediction? None of the above will matter and people everywhere will have to live with a permanent new mortal threat vector in their lives. My hope? Once these systems are routinely countered then we'll find a way to discourage or prevent their widespread use in the first place.

This is true of absolutely everything. If you want to you can argue against trying to control any hazardous substance or unethical weapon based on the argument that ‘the North Koreans won’t listen’. It gets the rest of the world nowhere.

I don’t think I expressed myself clearly enough. I fully agree that the best C-UAS is likely to be an autonomous C-UAS drone. My point is, if and when you can field such an effective C-UAS design that the enemy’s UAS are effectively nullified, you have stopped their autonomous attack drones from being deterministic weapons. The ‘do or die’ argument for unrestricted targeting of enemy soldiers, etc. has disappeared. C-UAS is now (arguably) the deterministic system since it basically grants the successful user the choice of how to prosecute the rest of the war. That is when it would be viable, imo, to eschew autonomous targeting of human beings and promote global abandonment of such an idea. Autonomous targeting of other enemy equipment (ships, aircraft, UGVs, etc.) would still be fine - that’s not the animalistic nightmare-inducing stuff. Autonomous targeting of individual people/crowds of people is what could be abandoned and there’s a chance the world might hurry to agree, for once.

This is only true until they are countered. The real power play in all this would be to focus almost entirely on an affordable and extremely effective C-UAS system. Once western armies can reliably counter an enemy’s drone fleet then they could take the moral high ground by eschewing autonomous human-hunting killbots (FPVs and autonomous targeting of enemy vehicles are still fair game) and driving a worldwide conversation about everyone else doing the same.

I think many countries were probably quite happy to sign up to nuclear non-proliferation given the difficulty and expense involved in setting up your own nuclear arsenal. Also, if anyone tried to breach non-proliferation treaties then there existed the legacy nuclear-armed powers who were able to carry out enforcement. There will be basically no significant cost/difficulty barrier to establishing an autonomous killer drone fleet, once the technology exists. That means any country will be able to do it almost at will, and, if they do, who would be able to stop them? I think it'd have to be someone with an even bigger fleet, no? Which means that, in this case, I don't think a treaty can work in the way we'd like it to. Maybe the real answer is to stop thinking about developing multi-layered C-UAS as a way to free up areas to manoeuvre in southern Ukraine and to start considering it a matter of humanitarian necessity.

Every time your autonomous drone tries to communicate, my C-UAS system is ready and willing to listen. I know we've been round the loop a few times but I, for one, enjoy thinking through this puzzle and reading others' thoughts as well. Keep it up.

Just riffing a little off what @sross112 was saying, in all these scenarios there has to be some thought given to where we think the UAV-war will find a balancing point. That means that c-UAS is the key. If we allow ourselves into the world of autonomy making EW all-but ineffective, there are a couple of options with regards to what form c-UAV ends up taking: Autonomous c-UAV drones ('fighter' drones) are effective against enemy UAS which fly above the treeline (or any other appreciable ground clutter) but everyone struggles to make them effective against those which are small and/or agile enough to travel amongst trees, hop over people's garden fences, etc. This world means that the compromising effect which UAVs are currently having on efforts to employ legacy systems (towed arty, MBTs, etc) will persist as long as the enemy have low-flying attack UAVs. This, I think, is the world most people on this thread are talking about and is the most likely to result in the most UAV-heavy future force compositions. Autonomous c-UAVs quickly gain sufficient sensor/AI levels to be effective against basically all enemy UAVs. You now have a world similar to the WW2 air war where the main battle is for air superiority and then your land forces can engage however they see best. At this point, once you have won air superiority, you ask yourself how best to attack the enemy: other drone designs? HIMARS? any old towed arty lying around? All those will have their pros/cons but you don't need to worry about the enemy UAS threat, at least, so some of them might remain on top tier TOEs. If ground-based anti-drone sniper units work, then see @The_Capt's thoughts, above - it will compromise UAVs but might do as much or more to compromise other legacy ground units, as well. If innovative forms of camo and concealment prove widely effective against future attack drone AI, that will also change the game. I think what all of the above probably hinges around most is the size, sensitivity and reliability of passive sensors of all relevant parts of the EM spectrum. If your autonomous UAVs can reliably see your chosen EM frequency at sufficient resolution, then I don't see what will stop AI getting us all to #2, above, pretty damned quickly (as well as probably enabling The_Capt's, nifty ground systems). Perhaps some of our resident subject-matter experts can opine as to whether there are any serious blockers to sensor design in certain parts of the spectrum which could then be exploited as UAV 'blind-spots' by both sides and result in a situation closer to #1?

I think I did start from the right place: the "effect" I proposed starting from was intended to mean any relevant effect, including your example of preventing a threat from reaching the battlefield. I admittedly did then leap to a projectile solution due to the nature of the discussion up to that point but I would wholly support your suggestion that the most powerful effects that a weapon designer may seek to achieve are often much further upstream than on the frontline.

Yeah, I went in a little heavy on the Stuka. When left alone to do their job they... did their job. They were good for hitting 100m targets relatively accurately, provided there was no modern or competent airborne opposition. As mentioned in a previous response my point was meant to be that the Stuka's reputation far outweighed its actual effectiveness relative to any other aircraft of its type and I believe that was due to the psychological impact its sirens (and partly its attack profile) had, early war.

Yup. Guns don't even have to be fired to have an effect.

Heh, I knew as soon as I posted that I wouldn't get away with the Ju-87 one. As much as I'd genuinely enjoy a discussion of the effectiveness of various WW2 aircraft designs I know you'll agree it doesn't belong on this thread. So, I will grant that it had military utility as far as any aircraft of its type had (although I think it lost its real utility long before it was phased out of service, even on the Eastern Front). Sooo, I should have written that the Ju-87 was militarily no more useful than a Dauntless or a D3A and was probably substantially less useful than a Typhoon or an F-series Fw-190 but, and this was my point, there's a reason the Stuka gained and retained such a terrifying reputation where those others didn't. It was the noise. Regarding V1s, absolutely they can't and shouldn't be compared to Stukas. My intention was to point out the effectiveness of the noise they made vs the destruction they caused and the fact that that led to their gaining an outsized reputation in the (certainly British) cultural consciousness. All this in the context of wondering why, for example, Russian terror-attack drones don't try something similar. Thanks for keeping me honest though - always appreciated.

Imo you're one order away from the truth, here: the hard requirement is the effect (including the type of effect). You'll probably need to transport an object to the target to cause the effect but that's not the start. Again, I think you're one dimension away: you want to "delay the collapse of the weapon's time and space option space" not necessarily as long as possible but at least until the point at which the target can no longer avoid being hit and there is therefore no longer a need to re-target. The rifle bullet is fine if fired from close range. The FPV drone is stuffed if the target drives away from it at 100km/h. What matters isn't the energy profile of the weapon system per se but its relationship to the intended target. If you start from an intended effect, you can decide what the best target will be and what the best type of effect would be (chemical, kinetic, phonic, electromagnetic, etc.). You can then work out the best way to apply that effect to that target (i.e. the type of warhead) in order to achieve the intended overall effect (I'll google synonyms for "effect" in a minute, don't worry). The mass, volume and fragility of the selected warhead will be the main things that dictate the achievable energy-time curves for your weapon system. Then you can start worrying about things such as launch signatures or changing trajectories post-launch and whether you can realistically do anything about those things. I've already written about the "as long as possible" point but you mention retaining energy here and that's important. Retaining energy is physically expensive and should always be seen as a compromising factor. All else being equal you want to retain as little energy as necessary after launch in order to achieve your desired effect at the target.

If we are trying to learn how to do war better, though, I don't think we can ignore the "soft and fuzzy" stuff at all. You have been among the first to remind people on this very thread that the world doesn't stop turning when the war ends and there are many very recent examples of it all going Pete Tong when people have forgotten that. You have to have a lasting relationship with the people you just finished fighting. How much harder is that going to be if you spent the whole war doing everything you could to terrorise them in ever more imaginitive ways; drilling right down into their amygdalas with screaming drones and running spider-mines to teach them an instinctive loathing of contact with your forces? As I said, I'm really surprised that some people don't do this stuff more. I would however advise caution if we were to think of doing it more. At least I would if we ever want to be welcomed anywhere as "liberators", again.