How Hot is Ukraine Gonna Get?

[cesmonkey]

 

[FancyCat]

Hungary has issued its response to Ukraine's block on oil. In short, the same money its been blocking....remains blocked! Bunch of idiots. 

Quote

“As long as this issue is not resolved by Ukraine, everyone should forget about the payment of the 6.5 billion euros of the European Peace Facility (EPF) compensation for arms transfers,” Foreign Minister Péter Szijjártó was quoted as saying by broadcaster ATV.

The EPF, created in 2021, operates as a cashback scheme that gives EU members refunds for sending munitions to other countries.

Since Russia’s invasion of Ukraine in 2022, member states that gave weapons to Ukraine can ask for compensation from this fund. But Hungary has been blocking the disbursement of the next tranche of EPF money for more than a year, citing various issues.

https://www.euractiv.com/section/global-europe/news/hungary-to-block-eu-funds-for-member-states-until-ukraine-allows-lukoil-transit

 

[The_Capt]

4 hours ago, holoween said:

And as said i dont disregard the drones for mining i just dont see them placing mines particularly far behind enemy lines. And i found it weird that you entirely forgot about arty deployed mines when they were quite a prominent feature in the winter offensive for closing breached minefields and placing new ones just ahead of assaults.

When did I forget about arty mines. Problem is numbers.  We never had that many systems back in the day and no one is really mass producing those types of systems, may change but considering we are having trouble keeping up on basic HE, FASCAM systems have to be getting rarer.  The advantage of drones is that they can basically take any mine - which there are still millions of - and deploy them 10 - 15 kms deep, no need for a cargo shell or special design.

I don’t believe for a second the MLRS or artillery are not maintenance intensive.  No large conventional system is hands free maintenance and logistics.  A drones is next to zero in comparison.  

[The_Capt]

3 hours ago, dan/california said:

 

^Actually...

^{Edit: And if you believe they don't have a fire and forget mode I have this really nice T-90 in my garage for sale, Cheap!}

Ok, now I want one with lasers and dazzle lights that hypnotize the enemy into thinking they are ducks…c’mon Internet….

[dan/california]

 

49 minutes ago, The_Capt said:

Ok, now I want one with lasers and dazzle lights that hypnotize the enemy into thinking they are ducks…c’mon Internet….

 

Quote

 

https://www.instructables.com/Duck-Cam-Decoy/

R/C Duck Decoy-Camera Hack

 

These have potential along the Dnipro.

[ArmouredTopHat]

 

https://breakingdefense.com/2024/07/swarm-wars-pentagon-holds-toughest-drone-defense-demo-to-date/

Man I would love to get the juicy details on this one, numerous systems being tested and while the details are sparse, apparently good results were obtained compared to previous showings. Key point is layered defence featuring soft and hard kill being more effective. The key word used was a maturing state of affairs for counter UAS.

I once again iterate the potential for counter UAS systems that might be being underestimated here in its ability. Very interesting that they specifically tested for drone swarms. 

Edited July 25 by ArmouredTopHat

[Carolus]

Mashovets on his telegram expressed worries about the expected manpower increase of the russian army in Ukraine to 690.000 men until the end of the year.

The UAF is being hard pressed on multiple front sectors. 

While the dynamics of the overall combat likely won't change, is anyone worried that this will enable RA to increase the frequency of their successes? 

They exhaust a specific position, often only held by three or four Ukrianians, over the course of days and then locally overwhelm. This leads to the small, gradual taking of territory that we observe, but also to the loss of this UAF team. 

I have the suspicion that the RA believes if they can just do "the same but more often" / in more locations they are able to out-attrite the UAF before internal regime stability or the economy goes down the gutter.

And something like a "Polish Brigade" with 1200 men will not make much of a difference here. 

Tl;dr even if time is not on Russia's side in the longterm, they seem to try really hard to reach the Ukrainian breaking point first and then sort out the aftermath. And the degree with which Western support is being either drip-fed or comes in irregular bursts is probably helping them to believe this idea can work.

Edited July 25 by Carolus

[Battlefront.com]

55 minutes ago, ArmouredTopHat said:

 

https://breakingdefense.com/2024/07/swarm-wars-pentagon-holds-toughest-drone-defense-demo-to-date/

Man I would love to get the juicy details on this one, numerous systems being tested and while the details are sparse, apparently good results were obtained compared to previous showings. Key point is layered defence featuring soft and hard kill being more effective. The key word used was a maturing state of affairs for counter UAS.

I once again iterate the potential for counter UAS systems that might be being underestimated here in its ability. Very interesting that they specifically tested for drone swarms. 

It is good to see some more stressful tests than what normally is seen by us in the public.  It is definitely what's needed and I do expect that each one of these tests will get us closer to having some form of viable C-UAS even if only for a limited subset of what needs protecting (e.g. facilities).

Some things to keep in mind, though.  First is this comment from the article:

Quote

Other than that broad brush, however, JCO officials pointedly declined to discuss how well the candidate counter-drone defenses did.

So we really don't know much of anything.  I've seen plenty of statements like this after tests by weapons that either never got produced, should never have been produced, or were stopped somewhere inbetween the two.  In fact, vague and positive statements that later turn out to be more BS than not is pretty normal.

Second, we don't know to the degree the effectiveness of any one system was.  We also don't know how that relates to the vague statement that the results were very good.  For example, if 5 of the systems failed to do much of anything, what was it that made the others successful?  If it was jamming radio signals... that's worrisome.  It would be like someone demonstrating a weapon that was great at sinking battleships just as battleships were all being retired.  The claims of effectiveness might be true, but it wouldn't be particularly useful.

Third, we have NO idea what the testing criteria was or how much/little it relates to real world situations.  Was it like the Rheinmetal PR video of their cannon shooting down a single wave of high flying drones in a nice, evenly spaced formation flying nice and slow straight towards the cannon?  Or were there multiple waves coming in from multiple points at different speeds, altitudes, angles, etc.?

Aside from the above, I think we can be very clear on four things:

1.  They chose perfect terrain for the C-UAS systems and the worst for drones.  This is fine for early stages of proof of concept, but it's very short of what these systems need to be capable of.  Until the test terrain is complex, the results of this particular test are academic only.

2.  Since this was a technical test they almost certainly didn't have the drones operated by an experienced "OPFOR" drone unit with carte blanche to do what they felt was best.  Instead, the test regime was almost certainly ridged and overseen by people who were not tasked with defeating the C-UAS.  Their most likely goal was to challenge the C-UAS, which is a much lower standard.

3.  Fourth, we've talked about how things like spoofers, chafe, jammers, etc. could be employed by the UAS in order to minimize the chances of being intercepted.  We know, for sure, that C-UAS used multiple layers, but did the attacking UAS?  Based on how testing at this stage usually works, I think we can be pretty sure that it was a lopsided test where one side (C-UAS) had all the advantages and the other side (UAS) was constrained by what it was able/allowed to do.

4.  It is not in the best interests of finding a C-UAS solution to set the bar too high (see above) because this discourages companies from trying.  Think about it.  If you've sunk $100m in venture capital into a C-UAS system and it scored well on a relatively simple test, then you might think it worth putting in another $100m to get it to the next level.  On the other hand, if you put in $100m into something that utterly failed, and I mean humiliatingly so, then maybe you don't put your next $100m into that system but instead abandon it and move onto something easier.

What I'm getting at here is that C-UAS systems are still quite far from being fielded and proven effective.  They are way behind where UAS currently is, not to mention where it is headed.

Steve

[Battlefront.com]

1 minute ago, Carolus said:

While the dynamics of the overall combat likely won't change, is anyone worried that this will enable RA to increase the frequency of their successes? 

Yes.  In fact, I have always been concerned about this since before the war started.  As a recap, I was one of the few people that correctly called Russia's fiasco in Ukraine, but I also thought that weight of numbers would eventually cause Ukraine's conventional defense options to collapse.  Obviously I was wrong about that, and I'm thankful for that!

Since Ukraine's early successes of holding back Russian advances and taking back some territory, I've maintained that there's only three ways this war can end on Ukraine's terms:

1.  Russia as a nation state collapses, at least to the extent of being able to maintain an offensive minded strategy.

2.  Russia's military collapses, at least to the extent of being able to maintain an offensive capability.

3.  Russia concludes the war is not worth it and seeks to get out of it before either a state or military collapse.

Since we don't see any of these three things happening any time soon, there is great risk to Ukraine that Russia will (as it is trying to do) wear them out.

Steve

[sburke]

Ukrainian forces down Russian drones with HIMARS in daring raid (msn.com)

 

Not sure why they said "shoots down", but it looks like they hit a high value target.

The Ukrainian Armed Forces executed an operation behind Russian lines in the Donetsk sector, managing to shoot down the latest Russian reconnaissance system, the 1K148 Jastreb-AW. This operation was conducted using the HIMARS system. 

The footage was recorded by the 14th UAV Regiment and shared on Telegram by the Head of the President's Office, Andriy Yermak.

This loss is significant for the Russians. The system was instrumental in detecting Ukrainian howitzers, mortars, and artillery systems such as HIMARS.

[ArmouredTopHat]

3 hours ago, Battlefront.com said:

It is good to see some more stressful tests than what normally is seen by us in the public.  It is definitely what's needed and I do expect that each one of these tests will get us closer to having some form of viable C-UAS even if only for a limited subset of what needs protecting (e.g. facilities).

This certainly looks a lot more comprehensive than a lot of the firing range tests I have seen in publicly available footage from companies and its good to see. There seems to be a genuine competition from these systems to determine the best ones for use.
 

3 hours ago, Battlefront.com said:

So we really don't know much of anything.  I've seen plenty of statements like this after tests by weapons that either never got produced, should never have been produced, or were stopped somewhere inbetween the two.  In fact, vague and positive statements that later turn out to be more BS than not is pretty normal.

Second, we don't know to the degree the effectiveness of any one system was.  We also don't know how that relates to the vague statement that the results were very good.  For example, if 5 of the systems failed to do much of anything, what was it that made the others successful?  If it was jamming radio signals... that's worrisome.  It would be like someone demonstrating a weapon that was great at sinking battleships just as battleships were all being retired.  The claims of effectiveness might be true, but it wouldn't be particularly useful.

Third, we have NO idea what the testing criteria was or how much/little it relates to real world situations.  Was it like the Rheinmetal PR video of their cannon shooting down a single wave of high flying drones in a nice, evenly spaced formation flying nice and slow straight towards the cannon?  Or were there multiple waves coming in from multiple points at different speeds, altitudes, angles, etc.?

Entirely fair observations to make, though I can understand why they are being pretty sparse with the results, the article itself mentions there is a lot of data crunching taking place right now. The comment that a layered system worked best does mesh with US practical experience against drone attacks in Syria though, and given this was testing between different competing systems instead of a demonstrator of one particular system, I would be willing to bet this has been a far more comprehensive test than most.

Certainly the West is very keen to find out what works best out of the options available due to the severity of the threat in question, this gives me hope in that they would be pretty stringent about determining what works best. The article points out the vendors were -not- told of what they would be facing up against specifically, which to me indicates that the tests were pretty serious and that the evaluation was less of a demonstration and more a serious actual effort to see what system worked best with unexpected threats. The fact they selected 9 systems from an initial 60 submitted says a lot about how serious they are in wanting to have the best defence designed so far, as is the comment that these systems performed much better than last year. It says a lot about how CUAS has evolved so much just in the last year alone in terms of maturity and proof of concept. The defence is catching up and it will be very interesting to see how it boils down overall. 

The testing itself involved drones from the bigger groups three Geran types drones to the small FPV group ones with simulated attacks, so all bases are covered, at least its suggest as such. The article also states specifically that these attacks were various and multi approach in nature in both angle and altitude, so on the surface of it it seems like a pretty comprehensive and realistic attempt to simulate attacks on a system. The article specifically mentions how they wanted to know how each system coped with numerous incoming threats and what was more efficient at processing the threat as well as disposal of it. This all to me sounds very encouraging in that they are looking for the right criteria. 

 

3 hours ago, Battlefront.com said:

Second, we don't know to the degree the effectiveness of any one system was.  We also don't know how that relates to the vague statement that the results were very good.  For example, if 5 of the systems failed to do much of anything, what was it that made the others successful?  If it was jamming radio signals... that's worrisome.  It would be like someone demonstrating a weapon that was great at sinking battleships just as battleships were all being retired.  The claims of effectiveness might be true, but it wouldn't be particularly useful.

From what I can ascertain from the article, each system was tested individually, so for them to say that results were very promising compared to previous testing indicates that all 8 at least worked to some degree. The take I got from the article was that most if not all these systems offered more than one means of defeating a drone. Presumably a combination of soft and hard kill. Certainly these multi-kill systems were noted to be far better in overall effect than a single response.

 

3 hours ago, Battlefront.com said:

Since this was a technical test they almost certainly didn't have the drones operated by an experienced "OPFOR" drone unit with carte blanche to do what they felt was best.  Instead, the test regime was almost certainly ridged and overseen by people who were not tasked with defeating the C-UAS.  Their most likely goal was to challenge the C-UAS, which is a much lower standard.

Another fair point, though we should also note that a swarm of 50 drones attacking in a short space of time is quite a bit more than what an experienced drone unit can typically throw at a system currently. The fact they are looking ahead at not just individual attacks but swarms is encouraging in that respect. I would be more sceptical if this was testing that involved one drone at a time. Its obviously early days and should a system be selected for potential service we can anticipate far more rigorous testing to that effect. 
 

3 hours ago, Battlefront.com said:

They chose perfect terrain for the C-UAS systems and the worst for drones.  This is fine for early stages of proof of concept, but it's very short of what these systems need to be capable of.  Until the test terrain is complex, the results of this particular test are academic only.

We dont know this for certain, for all we know the testing including placing the system in an area filled with LOS blockers to see how it coped. Probably best to wait until we know more before we conclude this was an 'easy mode' demonstration. 

 

 

3 hours ago, Battlefront.com said:

Fourth, we've talked about how things like spoofers, chafe, jammers, etc. could be employed by the UAS in order to minimize the chances of being intercepted.  We know, for sure, that C-UAS used multiple layers, but did the attacking UAS?  Based on how testing at this stage usually works, I think we can be pretty sure that it was a lopsided test where one side (C-UAS) had all the advantages and the other side (UAS) was constrained by what it was able/allowed to do.

Again a very fair point, but such capabilities from attacking drones are still very much a concept as well. We have already seen from the UA pov that AI targeting for instance is not currently fully viable even with its limited use so far (I recall something about it being quite expensive / tricky to implement it being mentioned here a few pages ago) Certainly drones could potentially do a lot more to spoof and fool counter systems, but then that does come at the cost of complexity and cost per platform for the attackers, not to mention things like size and weight increases that would only make them more detectable. It also constrains the biggest advantage of a small drone munition in that it would likely drive up its cost.
 

3 hours ago, Battlefront.com said:

It is not in the best interests of finding a C-UAS solution to set the bar too high (see above) because this discourages companies from trying.  Think about it.  If you've sunk $100m in venture capital into a C-UAS system and it scored well on a relatively simple test, then you might think it worth putting in another $100m to get it to the next level.  On the other hand, if you put in $100m into something that utterly failed, and I mean humiliatingly so, then maybe you don't put your next $100m into that system but instead abandon it and move onto something easier.

On the flip side, these competing companies know that this is a -very- important aspect of defence that everyone is very deeply interested in. In essence this might be one of the biggest defence sectors when it comes to demand going forward. The investment is probably more than worth the potential pay-out if they are selected, they already did well enough to be selected from literally dozens of potential systems as it stands. To put it mildly this is a giant prize valued in the billions when it comes to future contracts and you can expect these competitors to treat it very seriously to earn it, because any system selected from this initial testing is going to become very prominent in the defence sphere. 

 

3 hours ago, Battlefront.com said:

What I'm getting at here is that C-UAS systems are still quite far from being fielded and proven effective.  They are way behind where UAS currently is, not to mention where it is headed.

Absolutely. This is very much early days that hold a lot of maybes, though this sort of testing is clearly a step in the right direction that shows that the US at least is taking this very seriously, both with current capability and future capability of drone munitions. For all we know it could all be a waste of time, but at the same time it could be very potent at disrupting drone attacks, we just dont know yet.  I personally do wonder if we are seeing something akin to the 'happy time' experienced by Second world war Uboats, IE drones are enjoying a brief moment of unparalleled success against unprepared targets that can only offer improvised and adhoc responses currently before the defence catches up and suddenly something like an FPV drone is facing a lot more effective systems designed from the ground up to counter them. The future is scary and its very hard to say where we are going to end up.

What I found curious most of all was that none of the systems were directed energy types at all, which is very interesting. The systems tested included at least one vehicle mountable system as well, though its not clear if that means a dedicated chassis or the RWS style mountings we are seeing from some defence companies. If its the latter its rather key as that means you can put it on your traditional vehicles, which is big to say the least when it comes to individual vehicle protection against loitering munitions. 

Edited July 25 by ArmouredTopHat

[ArmouredTopHat]

 

[chrisl]

17 hours ago, Grigb said:

I'm not talking about tunnels. I'm talking about underground dugouts. Underground dugouts are bread and butter of the current war. 

Keeping most of the people in dispersed dugouts or beneath overhead/thermal cover can easily deceive casual observers, even satellites. Then you instruct a handful of people to move forrad at random intervals along irregular pathways to various forward dugouts. The casual observers will notice nothing but a few of guys here and there.

In a few days, there will be a few companies in forward dugouts, with satellite observershaving little idea what is going on.

And how do you attack from there?  If everybody is just hiding in dugouts and can't move in the open, aren't they just waiting to be attritted by an endless supply of cheap drones flying into the dugouts?  Depending on your opponent, there may be no such thing as "thermal cover" short of tunnels.  

That's fine for small defensive positions. But how do you feed them? Water?  Where does their trash go?  Where's their latrine? How does ammo come up?  A small team can spread their signs of existence around, especially in an already heavily impacted battlefield. A large force is going to be detectable and measurable by its secondary signs unless it's fully tunneled.

[Carolus]

Three Shahed drones about 12km deep in the Romanian border, claimed to be intercepted by F16. 

Might have been an error, might have been a test.

Romanian government is silent on the issue, so no official confirmation yet.

 

 

[chrisl]

15 hours ago, dan/california said:

You are making an excellent argument for being good at trench warfare under the all seeing eye. But if this is what you have to do to make a battalion level, maybe, attack? Then some combination of new technologies has changed the battlefield utterly.

This is more speculative, but I think you are underrating the ability of AI/machine learning to process surveillance data going forward. It seems entirely doable to have a computer system keep track of every detected movement. IF it can do that, it is trivial to have it keep track of how many people/vehicles/ect have moved into and out of a given area. This would cause this sort of of slow concentration to stand out pretty clearly. The systems may not be there yet, but it is certainly where they are trying to go. And this is vastly easier in a situation like Ukraines where there is just not much civilian activity in the most contested areas.

Every detected movement?

Try every object in the field of view.  Every rock.  Every footprint.  Every candy wrapper.  You don't even need AI for that.  Simple linear programming and some straightforward algorithms will show you a time series of changes over time with a single image.

"Hey, there are 3 new MRE wrappers in this section of image every ~4 hrs. And there are 3 new catholes in their latrine area every morning".  There are also extra footprints from this dugout to that dugout. From the thermal signature it looks like they were made in the last 2 hours"

[chrisl]

12 hours ago, The_Capt said:

See my post above.  Someone is going to mate Spike LR with a drone and welcome to hell.  The other way to go is to go smaller and put DPICM onto fully autonomous small drones in a swarm…essentially a smart cloud of cluster munitions.  Take that little monster and package it on a larger mother drone so you can get the distance.

Porque no los dos?

[The_Capt]

7 hours ago, Battlefront.com said:

It is good to see some more stressful tests than what normally is seen by us in the public.  It is definitely what's needed and I do expect that each one of these tests will get us closer to having some form of viable C-UAS even if only for a limited subset of what needs protecting (e.g. facilities).

Some things to keep in mind, though.  First is this comment from the article:

So we really don't know much of anything.  I've seen plenty of statements like this after tests by weapons that either never got produced, should never have been produced, or were stopped somewhere inbetween the two.  In fact, vague and positive statements that later turn out to be more BS than not is pretty normal.

Second, we don't know to the degree the effectiveness of any one system was.  We also don't know how that relates to the vague statement that the results were very good.  For example, if 5 of the systems failed to do much of anything, what was it that made the others successful?  If it was jamming radio signals... that's worrisome.  It would be like someone demonstrating a weapon that was great at sinking battleships just as battleships were all being retired.  The claims of effectiveness might be true, but it wouldn't be particularly useful.

Third, we have NO idea what the testing criteria was or how much/little it relates to real world situations.  Was it like the Rheinmetal PR video of their cannon shooting down a single wave of high flying drones in a nice, evenly spaced formation flying nice and slow straight towards the cannon?  Or were there multiple waves coming in from multiple points at different speeds, altitudes, angles, etc.?

Aside from the above, I think we can be very clear on four things:

1.  They chose perfect terrain for the C-UAS systems and the worst for drones.  This is fine for early stages of proof of concept, but it's very short of what these systems need to be capable of.  Until the test terrain is complex, the results of this particular test are academic only.

2.  Since this was a technical test they almost certainly didn't have the drones operated by an experienced "OPFOR" drone unit with carte blanche to do what they felt was best.  Instead, the test regime was almost certainly ridged and overseen by people who were not tasked with defeating the C-UAS.  Their most likely goal was to challenge the C-UAS, which is a much lower standard.

3.  Fourth, we've talked about how things like spoofers, chafe, jammers, etc. could be employed by the UAS in order to minimize the chances of being intercepted.  We know, for sure, that C-UAS used multiple layers, but did the attacking UAS?  Based on how testing at this stage usually works, I think we can be pretty sure that it was a lopsided test where one side (C-UAS) had all the advantages and the other side (UAS) was constrained by what it was able/allowed to do.

4.  It is not in the best interests of finding a C-UAS solution to set the bar too high (see above) because this discourages companies from trying.  Think about it.  If you've sunk $100m in venture capital into a C-UAS system and it scored well on a relatively simple test, then you might think it worth putting in another $100m to get it to the next level.  On the other hand, if you put in $100m into something that utterly failed, and I mean humiliatingly so, then maybe you don't put your next $100m into that system but instead abandon it and move onto something easier.

What I'm getting at here is that C-UAS systems are still quite far from being fielded and proven effective.  They are way behind where UAS currently is, not to mention where it is headed.

Steve

Looks like the tests were out in the desert…probably Yuma. [damn, Steve already picked up on this]  They are going to have to move to complex terrain to do any final testing.  My honest opinion is still that the best counter to a bunch of cheap autonomous strike drones is a bunch of autonomous CAP drones intercepting 2-3kms out.  Point Defence is last resort SHORAD in case something gets through.  Of course this will change ground manoeuvre fundamentally as we move away from DF centric firepower.  Next we will need UGVs out in a screen.  The next question is “what happens when two bubbles collide?”

BTW the level of OPSEC on this is an indicator just how seriously the US military is taking this.

Edited July 25 by The_Capt

[The_Capt]

3 hours ago, chrisl said:

Porque no los dos?

 

 

Edited July 25 by The_Capt

[ArmouredTopHat]

https://www.reddit.com/r/CombatFootage/comments/1ebq7gb/combat_operation_footage_from_the_outside_of_a/

Brief but once again a good display of how vehicles can withstand munitions via manoeuvre...and the almost casual shrugging off of an FPV munition. (Direct hit to the turret yet the thing still traverses just fine with no residue smoke or damage)

Interesting titbit of info from the comments below regarding the location of the camera. 

It's on top of the "doghouse" or the gunner's sight periscope housing. Originally that box contained the satellite antenna for the FBCB2 system, the second generation. Those systems were probably stripped from M2 provided to Ukraine because they're still fairly sensitive communication systems, and Ukraine wouldn't be able to fully use them without regularly getting cryptographic keys from the US.

You can see it in this picture, on the top/right (as you're looking at it), above the "windows" on top of the turret.
https://imgur.com/a/Yitli9t

[dan/california]

2 hours ago, The_Capt said:

BTW the level of OPSEC on this is an indicator just how seriously the US military is taking this.

And that, at least, is reassuring.

[A Canadian Cat]

Wow that was one badly written article.

Just one example of the poor writing. I suppose it could be a poor translation too. I should consider that.

9 hours ago, sburke said:

Not sure why they said "shoots down", but it looks like they hit a high value target.

Looks like they hit a launcher - which would be excellent.

I strongly suspect they did *not* bring a HIMARS system behind enemy lines to conduct that strike. I think they meant that the HIMARS hit something behind Russian lines.

 

[ArmouredTopHat]

It is good that the Russians remain incapable of properly sealing pockets and inflicting far more severe damage to units. 

Edited July 25 by ArmouredTopHat

[MikeyD]

SWARM WARS. I'd hate the be the rage safety officer at that anti-drone demonstration. Firing in all directions with various calibers at high angles. There's no telling where the rounds will eventually land.

[dan/california]

5 minutes ago, MikeyD said:

SWARM WARS. I'd hate the be the rage safety officer at that anti-drone demonstration. Firing in all directions with various calibers at high angles. There's no telling where the rounds will eventually land.

There is something to be said for a desert proving ground with essentially nothing but sand for ten or fifteen miles. 

[LongLeftFlank]

6 hours ago, chrisl said:

And how do you attack from there?  If everybody is just hiding in dugouts and can't move in the open, aren't they just waiting to be attritted by an endless supply of cheap drones flying into the dugouts?  Depending on your opponent, there may be no such thing as "thermal cover" short of tunnels.  

That's fine for small defensive positions. But how do you feed them? Water?  Where does their trash go?  Where's their latrine? How does ammo come up?  A small team can spread their signs of existence around, especially in an already heavily impacted battlefield. A large force is going to be detectable and measurable by its secondary signs unless it's fully tunneled.

 

6 hours ago, chrisl said:

Every detected movement?

Try every object in the field of view.  Every rock.  Every footprint.  Every candy wrapper.  You don't even need AI for that.  Simple linear programming and some straightforward algorithms will show you a time series of changes over time with a single image.

"Hey, there are 3 new MRE wrappers in this section of image every ~4 hrs. And there are 3 new catholes in their latrine area every morning".  There are also extra footprints from this dugout to that dugout. From the thermal signature it looks like they were made in the last 2 hours"

Yup, this is exactly the grim near future I expect.

Resulting in an uninhabitable no-mans land, what, dozens of kms deep? Hundreds of kms?