How Hot is Ukraine Gonna Get?

[cesmonkey]

WarGonzo says Ukraine is noticeably pulling back near Kupyansk:

https://t.me/wargonzo/14305

Quote

️Important️The forces of the Russian Armed Forces advanced to Kupyansk️

Our sources on the ground confirm reports that the Armed Forces of Ukraine have left some of their oporniks in the Kupyansk direction. The military formations of the western group are carefully advancing after the retreating enemy.

The retreat of the Armed Forces of Ukraine may be part of a plan to lure our group from their positions in order to organize subsequent flank attacks.

However, we can really talk about the decision of the Armed Forces of Ukraine to change the front line and transfer the enemy's defense lines to a natural barrier - the Oskol River. In any case, the retreat of the Armed Forces of Ukraine is the result of the active work of our soldiers and their commanders in this direction.

 

[kevinkin]

Some bigger picture items in this report:

https://news.yahoo.com/ukraine-russia-war-live-explosion-042305134.html?fr=yhssrp_catchall

For example:

Putin ‘considers coming face to face with world leaders’ at G20 summit

New Wagner ‘tent city’ being built 15 miles from Belarus-Ukraine border, Kyiv group claims

Fear of tech ‘brain drain’ prevents Russia from seizing Yandex for now - sources

Ukraine's navy announces Black Sea humanitarian corridor but says Russian threat remains

[kimbosbread]

3 hours ago, Maciej Zwolinski said:

On the other hand, if the UKR stopped those penny packet attacks for some time and instead gathered say 100 cheap kamikaze drones to be launched as a single go e.g. at the Feodosia airfield, that could conceivably saturate the Russian defences and smash e.g. 20+helicopters.

First big problem is payload- can't use GPS, can't use video link to operator, good INS is too expensive etc. That's why I'm a big fan of autonomous optical guidance in conjunction with degraded GPS + poverty INS on a chip. You tell the munition go to the airfield- if you fly over this way, it should be there in 2 hours, and then hit whatever helicopter is not already on fire. That's completely doeable with a few motivated electrical and software engineers.
 

1 hour ago, Battlefront.com said:

Coming up with an adequate substitute for Western rocket based PGMs is a tall order.  Not without direct Western assistance can I see them being able to do anything like that.  So the best they apparently can produce are larger drones, which do not pack the amount of explosives that are necessary to put military and road infrastructure out of commission.  They also can be, and are, shot down.

As you point out, second problem is payload. Taking out a helicopter, or a truck, or even an individual soldier requires only a small warhead. Taking the bridge out of action won't work even with 1000x 2kg warheads.

Looking forward to South China Sea, I'm not convinced the USMC's tomahawk on an atv is the right choice, but I don't have a good sense for what the minimum size warhead is to sink or damage a ship. I think smaller, cheaper loitering munitions with a thermobaric warhead that would punch through the bridge roof or fly into the funnel or helicopter hangard would completely mess up a smaller warship or transport, but my only experience with watercraft is kayaks and 14 foot racing sailboats!

[kevinkin]

10 minutes ago, kimbosbread said:

I'm not convinced the USMC's tomahawk on an atv is the right choice

The first thing to consider is getting a system through the air defense the ship has in place organically. The fast the better I think. However, tiny might work especially with jamming. The naval does have this as a major weapon to use:

The AGM-158C LRASM (Long Range Anti-Ship Missile) is a stealth air launch anti-ship cruise missile developed for the United States Air Force and United States Navy by the Defense Advanced Research Projects Agency (DARPA).[5] The LRASM was intended to pioneer more sophisticated autonomous targeting capabilities than the U.S. Navy's current Harpoon anti-ship missile, which has been in service since 1977. Warhead: WDU-42/B HE blast fragmentation penetrator
Warhead weight: 1,000 lb (453.6 kg)

But for one alone to get passed the air defense would be lucky without a coordinated attack. Loitering might be too slow. But on the topic of unmanned, long range surface drones would be a real PIA since they are hard to detect. The US Navy has a close eye on those from the war I bet. 

[Maciej Zwolinski]

18 minutes ago, kevinkin said:

The first thing to consider is getting a system through the air defense the ship has in place organically. The fast the better I think.

I do not think the US Navy has ever had fast ASMs. Ever since the first variants of Harpoon and Tomahawk they have  gone for low altitude, low RCS and overwhelming numbers attack style.

[kevinkin]

25 minutes ago, Maciej Zwolinski said:

overwhelming numbers attack style

Hughes salvo model. Actually come into play now with strategic drone strike from the air. The old captain probably never considered unmanned stuff. 

[kimbosbread]

I’m not convinced long range surface drones are good for anything but attacking fixed infrastructure or ships in constrained areas like near a harbor (which is still a big deal). You need to know where the target is, and then wait 5+ hours for the USV to get there and then hopefully the target is still there. If the USV is autonomous and can attack targets of opportunity along the way or near the end if the desired target is not available, that’s better of course.

The money maker for the Taiwan Straight is loitering underwater munitions, ie smart mines. Sure, China can deny access to the straight, but it’s easy for Taiwan and the US to also deny access. It’s not clear to me there’s any counter at all to mobile sufficiently smart mines (smart enough to reliably distinguish a real ship in the AOA vs decoys). Way cheaper to build than ships, impossible to detect, and enough explosive to blow up a ship.

[kevinkin]

28 minutes ago, kimbosbread said:

You need to know where the target is, and then wait 5+ hours for the USV to get there and then hopefully the target is still there.

Not there today, so this will require innovation for sure. 

[cesmonkey]

Podcast episode interviewing John Nagl about his paper:

https://press.armywarcollege.edu/parameters/vol53/iss3/
A Call to Action: Lessons from Ukraine for the Future Force
Katie Crombe and John A. Nagl

 

[dan/california]

10 hours ago, Teufel said:

This is not intended as sarcastic smart *** one liner, but doesn’t the saying go “offense is the best type of defense”?

From previous rants of Russian disinformation and propaganda; Russians are pushing now, as in timing, to relieve pressure further south. Think they are in reality full of smelly hot gases. Thus lack actual capabilities to overrun Ukrainians and make significant push towards capturing Lyman/Izium.

So they do what they do best, show their Sovjet inherent incompetence of conducting effective offensive operations. And you said it yourself, resorting to “shelling the sh*t out of it”.

Edit: We discussed the Sovjet doctrine of attritional defensive strategies before. Hold, retreat, artillery cover on previous positions and counterattack, on and on and on. Only problem for Russians now is that Ukrainians are doing the same thing.

When Russians attack, they retreat, Russians set in artillery fire, revealing positions, and Ukrainians answer with actually accurate counter artillery.

Elmar nailed it on the previous page;  “The Russian strategic reserves of stupidity are not at any risk of running out, it seems.”

Edit 2: Speaking of Russian propaganda, this guy is legendary; “The necessary reports are required from you in the universe of military leadership, and if reality does not correspond to it, these are problems of reality.”

 

The War translated thing is worth the time to read in full.

Edited August 10, 2023 by dan/california

[Vet 0369]

On 8/8/2023 at 11:54 PM, Battlefront.com said:

I hate lobster almost as much as I hate having to wade through repeated instances of feather ruffling over something that adds nothing of value to this thread.

Steve

LOL! We share that dislike of lobster. After my sister and I had polio in 1955, our Father worked two jobs just to pay the medical bills. He also had a small boat and some lobster pots, so we ate lobster all the time. My Mother told me that one time I came home from elementary school and asked “why do I have to have lobster sandwiches all the time, why can’t I have peanut butter and jelly like the the other kids?” I will eat it, but only if I’m a guest and someone serves it to me.

Massachusetts actually had a law that prisoners and servants couldn’t be fed lobsters more than a certain number of times per week!

 

[Phil003]

4 hours ago, kimbosbread said:

First big problem is payload- can't use GPS, can't use video link to operator, good INS is too expensive etc. That's why I'm a big fan of autonomous optical guidance in conjunction with degraded GPS + poverty INS on a chip. You tell the munition go to the airfield- if you fly over this way, it should be there in 2 hours, and then hit whatever helicopter is not already on fire. That's completely doeable with a few motivated electrical and software engineers.

I don't think this would be a feasible option. I was working on autonomous driving for a while, around when the hype started, and I am still in the automotive sector and follow the topic more or less for different reasons (it seems engineers 'coming out' is the current theme of this thread  ) Based on what I see, I am pretty convinced -even though Elon Musk repeats every year, that full autonomous driving is just 1 year away- that considering the current state of the 'AIs', you can't trust them to make reliable/safe decisions (without human supervision) in an uncontrolled environment, like driving in a city, or selecting targets in the chaos of combat, etc.
So if you could make 100% sure that the drone is over the airfield when you allow it to select a target, then sure you can give it a try, and the worst thing that can happen that it will strike the helicopter statue in front of the base instead of a real helicopter, or something like that (here I am making the reasonable assumption that Russian pilots don't have 'take your child to work' days during wartime), but I guess, you can't guarantee close to 100% sure location information under all circumstances if you want to keep the drone cheap, so it might start looking for targets above a school, and pick a playground piece resembling to a helicopter to hit...
And even if this plan would work once (which is not impossible), when Russians realize what's going on, the next time they could easily fool the drones e.g. by putting tents over the helicopters, or make their shape unusual with some wooden attachments, or by anything that would confuse the image recognition, and for the best result they can combine this with deploying decoy 'dummy' helicopters a little bit away from the real ones.
So, basically I agree that something like this is possible to be done, but I would estimate that it would take a huge amount of development effort to make it more or less work, let's say with 90% reliability, but you have the problem of the risk coming from the remaining 10% which can result, in worst case, in killing a lot of civilians, and once the element of surprise fades, it is quite easy to fool the drones, so I would guess it just doesn't worth the effort and the risk.
I am aware that similar ideas, such as using 'AIs' to make critical decisions during combat, like target selection, were popping up on this thread repeatedly, and to be honest I am a bit puzzled about the confidence. So I never worked in the defense industry (though in case a CTO or similar is following this thread, I would like to state that currently I would be interested to do so ), but I am sure that the core of the problem is the same as in case of autonomous driving, you can make it work pretty good, but pretty good is not good enough when a single mistake can kill somebody (in case of military application, kill somebody you didn't intend to kill). Also, so far whenever I read about military autonomous systems which seemed reasonable close to actually being fielded (so not about experimental systems or prototypes), these systems are either performing non critical tasks, so definitely they are not firing weapons on their own, or they work under human supervision, so e.g. proposing targets, but a human operator makes the final decision (which for me seems to be the perfect setup, combining the capabilities and speed of the 'AI' with the (in this case still better) reliability of humans) So my feeling is that engineers in the defense industry are also quite cautious about the topic.

[acrashb]

From the Economist:

"The war in Ukraine shows how technology is changing the battlefield
But mass still counts, argues Shashank Joshi in the first of seven chapters of a special report on the future of warfare"

Another useful analysis of what this war means for the next war. Shashank Joshi is the defence editor for the Economist, previously a think-tanker at RUSI.

https://www.economist.com/special-report/2023/07/03/the-war-in-ukraine-shows-how-technology-is-changing-the-battlefield

I like a lot of it, but here's one quote: “You can’t cyber your way across a river.”

It's a paywall.  Recommended if you know how to bypass or just have subscription (well worthwhile).

 

[Vet 0369]

16 hours ago, Kinophile said:

Such an Engineer type of post.

So cute, in confusing not-cute way.

Question: How can you tell an extroverted Engineer from an introverted Engineer?

 

 

Answer: An extroverted Engineer stares at “your” shoes when talking to you!

[Maciej Zwolinski]

13 minutes ago, acrashb said:

Another useful analysis of what this war means for the next war. Shashank Joshi is the defence editor for the Economist, previously a think-tanker at RUSI.

A "think-tanker"? Is that rhyming slang?

[Haiduk]

One of memes after explosion on factory in Sergiev Posad ) 

The writing "Budangamer" )

Edited August 10, 2023 by Haiduk

[kevinkin]

https://www.washingtonpost.com/world/2023/08/10/ukraine-national-mood-counteroffensive-gloom/

Slow counteroffensive darkens mood in Ukraine

Throughout, Ukrainian officials and their western partners hyped up a coming counteroffensive — one that, buoyed by a flood of new weapons and training, they hoped would turn the tide of the war.

But two months after Ukraine went on the attack, with little visible progress on the front and a relentless, bloody summer across the country, the narrative of unity and endless perseverance has begun to fray.

While the story makes you angry want to cry, I have to hand it to the higher ups at the Post for publishing it. War on this scale is not clinical. I am interested in who pick's up on this piece across media outlets. Maybe it will drag the DC elites off the beach to answer tough questions with something other than boiler plate answers. 

[Haiduk]

18 hours ago, Fenris said:

And these have been posted in the same thread, go-pro footage of a river raid.  Can't be sure it's tehs same action, the text claims it is I think.

 

No, this is recent action, I posted other FPV video of this action with withrawing of boats under fire

Here is videos from Kozachi Laheri:

Russian soldiers flee in panic to the village from own positions on western outskirts. The long building is vegetable storage on western outskirt of Kozachi Laheri.

Other two videos:

1. Ukrainian MLRS RM-70 "Vampire" works at Russion positions likely before landing

2. Presumably the place of ambush of major Tomov recons in 4 km west from the village - knocked out BTR-82A

Yesterday Russian TV staged a reportage about "no Ukrainian troops in Kozachi Laheri, all is calm" - the journalist makes own stand-up on background of village's House of culture and on next day this building already damaged. 

According to Russian milblogger "13th" during yestreday clashes and ambush, Russian troops lost 25 KIA and MIA. UKR account 777UA, close to SOF, claimed our troops had only several wounded. 

IN present time still unknown either UKR troops abandoned Kozachi Laheri area or not, because today next Ru TG message appeared:

The battle is continuing, reportedly AFU already came to the road Chelburda - Kozachi Laheri. At the same time they bypassing Kozachi Laheri from south  and seriously threaten with encirclement our forces, which deployed in this village. 

This can be:

1. True

2. Hysteria and hype of Russian milblloggers

3. UKR PsyOps

4. Russian Psy Ops (old trick with fiction UKR advances, which heroically have been repelling by brave Russian troops)

Edited August 10, 2023 by Haiduk

[Haiduk]

Damaged by UKR drone SIG tanker. You can see a hole in the hull

 

[dan/california]

5 minutes ago, Haiduk said:

This can be:

1. True

2. Hysteria and hype of Russian milblloggers

3. UKR PsyOps

4. Russian Psy Ops (old trick with fiction UKR advances, which heroically have been repelling by brave Russian troops)

This should be attached to everything based on a Russian milblogger's reporting.

 

[Haiduk]

This night Moscow again was under attack of UKR drones. Despite all where intercepted, two Moscow airports Domodedovo and Vnukovo were closed for a long time. One explosion was in Domodedovo district, but city authorities claimed this was explosion of gas baloon in car-service faciity.

Because of this Il-76 from Rostov-on-Don with wounded, who have to be moved to Moscow was delayed and almost all day stood on local airfield. Wounded soldiers lay in the sun many hours without food and water. 

Also, reportedly UKR Mugin-5 PRO drones attacked Crimea, but Russians claimed all were shot down or supressed by EW. Though, strange smoke was filmed after this is the area of Novofedorivka airfield. Local authorities claimed this is just "grass burning"

[Der Zeitgeist]

Various German news media are reporting that providing KEPD Taurus to Ukraine is now seriously being considered by the German government.

One question seems to be the option of "geoblocking" Russian territory for the missiles, which is currently discussed with the German industry.

If this is true, this obviously raises the question if this includes Crimea. Have there been any confirmed Storm Shadow strikes on Crimean territory so far? I'm wondering what the "unofficial", "internal" definition of Russian territory for Western governments is right now...

Edited August 10, 2023 by Der Zeitgeist

[Haiduk]

6 minutes ago, Der Zeitgeist said:

Have there been any confirmed Storm Shadow strikes on Crimean territory so far?

Yes. Bridges, which were hit on past week for example

[kimbosbread]

2 hours ago, Phil003 said:

I don't think this would be a feasible option. I was working on autonomous driving for a while, around when the hype started, and I am still in the automotive sector and follow the topic more or less for different reasons (it seems engineers 'coming out' is the current theme of this thread  ) Based on what I see, I am pretty convinced -even though Elon Musk repeats every year, that full autonomous driving is just 1 year away- that considering the current state of the 'AIs', you can't trust them to make reliable/safe decisions (without human supervision) in an uncontrolled environment, like driving in a city, or selecting targets in the chaos of combat, etc.
So if you could make 100% sure that the drone is over the airfield when you allow it to select a target, then sure you can give it a try…

Autonomous flying is way easier than autonomous flying- we already have this, and it used every day- autopilot. Obviously there is reliance on GPS and INS, but it’s way different than being on a road and having to worry about obstacles and traffic. If GPS is jammed, poverty INS and simple airspeed calculations and a compass will get you close to where you need to go- especially if you fly higher up on a clear day, or just in straight line. And then simple image recognition to look for tarmac/flying vehicles. If you can get to the target 80% of the time, that’s great. No AI needed.

If you can’t make it to an airflield, ditch on an empty road and make a hole it in it and hope someone gets a flat tire.

In terms of target selection, if Russia has kids at a military airport in a war, I don’t think anybody is going to complain if they get killed. Maybe amnesty, but they have shown their true colors of course.

Detecting helicopters vs decoys, if you have an even halfway decent onboard camera and the equivalent of an iphone CPU, that’s plenty of horsepower to do image recognition. You don’t need an LLM or deep learning or anything crazy. Again, no AI needed. Same goes for detecting an airfield.

[kevinkin]

https://rusi.org/explore-our-research/publications/commentary/ukraine-war-has-found-machinery-western-governments-wanting

Article forwards the idea that inaction early in 2022 meant that Ukraine could not deliver a killing blow when the RA was in disarray over the winter. Ok, Monday morning quarter backing for sure. It does not take into account the state of the UA and it's ability to use the assistance donated if it arrived as the author wished. However:

Culturally, Western governments have spent decades writing long-term strategies and managing small-scale, short-term crises like terrorist attacks. It appears the institutional memory of how to cohere the operational level of war has atrophied. This malady is correctable, but only if we can acknowledge that there is a problem to be addressed.

So, the US and NATO were caught flat footed? The shift of US emphasis to the Pacific left Ukraine high and dry? And the response was/is the same old same old while Ukraine is in ruin? 

 

 

Edited August 10, 2023 by kevinkin