How Hot is Ukraine Gonna Get?

[The_Capt]

8 minutes ago, Battlefront.com said:

More footage of the failed Russian attack in the Kupyansk area yesterday:

 

Damn it is hard to tell what is doing the damage in these videos.  I think I see arty, mines and maybe an ATGM strike.

[hcrof]

25 minutes ago, Blazing 88's said:

Snagged from winSPMBT forum:

https://breakingdefense.com/2023/11/army-gets-its-hands-on-new-high-power-microwave-prototype-from-epirus/

WASHINGTON —The US Army has accepted delivery of Epirus’s first high-power microwave prototype for a new developmental initiative aimed at protecting soldiers and facilities from swarms of aerial drones, the company announced today.

The delivery marks the first of four prototypes derived from the company’s Leonidas counter-unmanned aircraft systems (cUAS) system that Epirus owes the service after inking a nearly three-year, $66.1 million contract in December 2022 for the Indirect Fire Protection Capability–High-Power Microwave (IFPC-HPM) initiative. A company spokesman told Breaking Defense the plan is to now deliver the second directed energy weapon to the Army’s Rapid Capabilities and Critical Technologies Office (RCCTO) by the end of December and complete delivery of the full platoon of four IFPC-HPMs early next year.

 

 

Do you know what it's costs and capabilities are? 

Seems pretty big and high tech for a drone killer. Maybe good for ultra high value targets but not general use!

[acrashb]

For general interest, "Timelapse of every battle in history".  It should say every _recorded_ battle, but is still interesting.  And the final frame showing all of them at the same time makes Europe look truly dangerous.
 

 

[The_Capt]

9 minutes ago, hcrof said:

Do you know what it's costs and capabilities are? 

Seems pretty big and high tech for a drone killer. Maybe good for ultra high value targets but not general use!

Big microwave oven trying to bake the sky goes boom.  Good for point defence against asymmetric attack.  Suicide in a conventional war.  A lot of this technology sprang up when ISIL started messing with UAS about 5 years ago.  It was built for a very different environment.

[The_Capt]

13 minutes ago, acrashb said:

For general interest, "Timelapse of every battle in history".  It should say every _recorded_ battle, but is still interesting.  And the final frame showing all of them at the same time makes Europe look truly dangerous.
 

 

A little Eurasian centric, but then again these were the people "recording history".  If (and it is a big if) I ever do a PhD, it will be on pre-historic warfare.  That map would see a massive mist of small lights that was the background warfare of pre-civilization.  Also there were very likely some very big battles in North America pre-contact...but of course no one wrote them down.

[Battlefront.com]

3 minutes ago, The_Capt said:

Big microwave oven trying to bake the sky goes boom.  Good for point defence against asymmetric attack.  Suicide in a conventional war.  A lot of this technology sprang up when ISIL started messing with UAS about 5 years ago.  It was built for a very different environment.

My first awareness of drone threats were the strikes on Saudi oil refinery infrastructure in 2019.  That was a real eye opener for me.

https://en.wikipedia.org/wiki/Abqaiq–Khurais_attack#Domestic_reactions

Steve

[Blazing 88's]

20 minutes ago, hcrof said:

Do you know what it's costs and capabilities are? 

Seems pretty big and high tech for a drone killer. Maybe good for ultra high value targets but not general use!

No idea, there are numbers in the article for the contract signed, which was 66+ million.

I added a related link to my previous post you may find interesting as well:

RELATED: To counter drones, Army seeks layers rather than ‘silver bullets’
https://breakingdefense.com/2022/10/to-counter-drones-army-seeks-layers-rather-than-silver-bullets/

[Kinophile]

11 minutes ago, The_Capt said:

A little Eurasian centric, but then again these were the people "recording history".  If (and it is a big if) I ever do a PhD, it will be on pre-historic warfare.  That map would see a massive mist of small lights that was the background warfare of pre-civilization.  Also there were very likely some very big battles in North America pre-contact...but of course no one wrote them down.

North & South America needs its Tollense River equivalent. 

[The_Capt]

12 minutes ago, Blazing 88's said:

No idea, there are numbers in the article for the contract signed, which was 66+ million.

I added a related link to my previous post you may find interesting as well:

RELATED: To counter drones, Army seeks layers rather than ‘silver bullets’
https://breakingdefense.com/2022/10/to-counter-drones-army-seeks-layers-rather-than-silver-bullets/

"Gainey noted that the electronic systems work “pretty good” against smaller unmanned systems, but he has previously said that the rise of autonomy in unmanned operations has limited the effectiveness of EW as “now you’re not cutting a link.”

This is driving adversary R&D in full autonomy.  They also do not have legal or policy regimes pushing back on this.

[Harmon Rabb]

Quo the Marlon Brando impressions, saying something about sleeping with the fishes.

Edited November 2, 2023 by Harmon Rabb

[Fernando]

45 minutes ago, acrashb said:

For general interest, "Timelapse of every battle in history".  It should say every _recorded_ battle, but is still interesting.  And the final frame showing all of them at the same time makes Europe look truly dangerous.
 

 

Hmmm, all battles fought in America from 1492  to 1700 are not shown.

[Battlefront.com]

52 minutes ago, The_Capt said:

"Gainey noted that the electronic systems work “pretty good” against smaller unmanned systems, but he has previously said that the rise of autonomy in unmanned operations has limited the effectiveness of EW as “now you’re not cutting a link.”

This is driving adversary R&D in full autonomy.  They also do not have legal or policy regimes pushing back on this.

Automation is yet another thing that needs to be broken down into sub components.

Navigational autonomy has existed since WW2 in some sense (I'm thinking of the V-1) but certainly since the early 1980s (Tomahawk).  This allows the weapon to navigate to a designated target without reliance upon Humans or disruptable navigation aids (GPS).

Targeting autonomy comes in two forms; smart and dumb.  Dumb is just something that is flung with no guidance, smart involves some sort of sensors that guide the munition to a target.  Cluster munitions, for example, started out being dumb but have gained capabilities to be guided by light and/or heat.

When you think of it, we've already had these technologies in use.  A 1990s era Tomahawk could be armed with 1990s era "smart" munitions normally packed into cluster bombs.  It would be a fully autonomous "smart" weapon in that it could get to a specific point on the map and target objects with a particular signature.

We've moved away from this sort of stuff lately because adding a Human back into the loop provides a LOT of benefits.  And it would have been incorporated into V-1s and Tomahawks back in the day if the technology had been available. 

Now that "Human in the loop" and satellite guidance are being countered it's sensible to look at improving the "smarts" to the extent that a drone can get to and find its own targets so as to strike them precisely using advanced sensors and AI in combination.  This is what gives people, smart people at least, cause for concern.

One thing that could help bridge the gap until we have AI we can "trust" is to revisit the older types of autonomous systems, add better sensors, and call it good enough for now.  Think a drone with LIDAR and a range of targeting sensors that can reasonably identify whatever it is tasked with destroying.  It wouldn't be able to go after Humans unless authorized before launch, but there are circumstances that would be perfectly fine.  For example launching the drone against an enemy training base or a frontline position.

Just thinking that full autonomy doesn't necessarily have to be the super scary stuff that we have every right to be super scared about.

Steve

[Beleg85]

Opinion about Avdiivka situation from Butusov:

[LongLeftFlank]

2 hours ago, The_Capt said:

If (and it is a big if) I ever do a PhD, it will be on pre-historic warfare.  That map would see a massive mist of small lights that was the background warfare of pre-civilization.

 

 

[Lethaface]

1 hour ago, Beleg85 said:

Opinion about Avdiivka situation from Butusov:

interesting opinion, thanks.

[Probus]

If you launch a swarm of drones, wouldn't it make sense to include some autonomous drones that target ECM sources with HARM-like missiles?  I know these missiles are pretty large for a small drone, but a new short range HARM might fit on a medium size drone.

[akd]

More Avdiivka insanity (Oct. 10):

 

Edited November 2, 2023 by akd

[Harmon Rabb]

 

[chrisl]

2 hours ago, Battlefront.com said:

Automation is yet another thing that needs to be broken down into sub components.

Navigational autonomy has existed since WW2 in some sense (I'm thinking of the V-1) but certainly since the early 1980s (Tomahawk).  This allows the weapon to navigate to a designated target without reliance upon Humans or disruptable navigation aids (GPS).

Targeting autonomy comes in two forms; smart and dumb.  Dumb is just something that is flung with no guidance, smart involves some sort of sensors that guide the munition to a target.  Cluster munitions, for example, started out being dumb but have gained capabilities to be guided by light and/or heat.

When you think of it, we've already had these technologies in use.  A 1990s era Tomahawk could be armed with 1990s era "smart" munitions normally packed into cluster bombs.  It would be a fully autonomous "smart" weapon in that it could get to a specific point on the map and target objects with a particular signature.

We've moved away from this sort of stuff lately because adding a Human back into the loop provides a LOT of benefits.  And it would have been incorporated into V-1s and Tomahawks back in the day if the technology had been available. 

Now that "Human in the loop" and satellite guidance are being countered it's sensible to look at improving the "smarts" to the extent that a drone can get to and find its own targets so as to strike them precisely using advanced sensors and AI in combination.  This is what gives people, smart people at least, cause for concern.

One thing that could help bridge the gap until we have AI we can "trust" is to revisit the older types of autonomous systems, add better sensors, and call it good enough for now.  Think a drone with LIDAR and a range of targeting sensors that can reasonably identify whatever it is tasked with destroying.  It wouldn't be able to go after Humans unless authorized before launch, but there are circumstances that would be perfectly fine.  For example launching the drone against an enemy training base or a frontline position.

Just thinking that full autonomy doesn't necessarily have to be the super scary stuff that we have every right to be super scared about.

Steve

Those older systems are really automation, rather than autonomy.  The systems are given explicit instructions on route and destination and they follow them based on their navigational sensors (dead reckoning for the older ones, GPS and maybe terrain relative nav and star trackers for some more recent ones).  They're not making decisions about anything other than where to turn, and those are based on either something really simple (clock time and airspeed or groundspeed) or something slightly more complex like error signal relative to external nav information (stars, GPS, terrain maps).  They might have an IR sensor that turns on at the end so that they get into the general neighborhood of something big and hot, then target the first hot thing (like a tank or a ship) that they see.  They're not really making independent decisions.  These systems don't need trust - they're point and shoot, where the pointing is a little fancier than LOS.

Autonomy is more mission oriented and you give a thing some general goals and it makes some sort of decisions based on a merit function: "I see a TOS and a T-72, I'll blow up the TOS", or "I see a bunch of vehicles, and based on some fuzzy training I have, I'm going to take out that one that looks like a command vehicle".  These systems need trust (or a human in the loop) because you're just sending them out on their own and don't want them to decide to specifically target civilians or friendlies out of all the stuff they see.  And what you describe is one of the first levels of autonomy that we use even for non-deadly things - let the robot sift through enormous amounts of data that may not be familiar and present what's likely the most interesting thing to the human operator for evaluation.  It typically also will send other samples (at regular intervals, or random, or from particular clusters) as a secondary check that it's doing the right thing and not missing things.

 

[kimbosbread]

On 10/31/2023 at 1:07 PM, The_Capt said:

Gotta be honest. Kind of where my head goes.  Big problem with energy density here.  A battle suit that basically makes the individual soldier the platform would solve a lot of this…

But how do you power the damn thing?  An exo-suit with armor will have a lot of weight and the only thing with enough energy to power one is fossil fuels, which is really problematic for many reasons.  So we would need something that can meet or exceed existing fuel energy density to power these things. 

I think we are closer to being able to rely on at least some battery power than is commonly accepted. V1 powered armor is likely just mobility, where the soldier can now jog at 10kmph with a full combat load for 1 hour without destroying their joints. Let’s say takes 1kwh. If the solider is packing along a few kg of tesla-style batteries, that starts look less sci-fi and more real.

I think the major challenge is engineering a suitable actuator (ie artificial muscle) for this suit.

18 hours ago, cyrano01 said:

I was starting to think something similar but I can't help wonder about the likely cost and cost/effectiveness of MI troopers. At what point do the mobility-suited, infantry soldiers start to resemble one-man tankettes for practical purposes?

It’s all about the economics. If 10 soliders cost as much as one tank, but each one has similar destructive capability by themselves relative to the tank, then things are ok.

16 hours ago, The_Capt said:

Full body armour might be a start but an exco-suit would need to be able to carry the weight and still allow for movement. 

Yeah we have to differentiate between the various benefits of powered armor, and how much they cost. For example, improved mobility, vs more weight, vs better camouflage, etc.

14 hours ago, dan/california said:

Watling's book…

Ugh I’ve only finished part 1 and a week or two of 12 hour days has put a halt to more reading.

[Battlefront.com]

39 minutes ago, akd said:

More Avdiivka insanity:

 

Holy crow!  That's a lot of vehicles!  Looks like they had picked off some from a previous attempt, but the video starts with at least one burning.  The second one that got nailed in the treeline appears to have held up the rest, so it was just charge of two vehicles.  I have to say, even after seeing hundreds of Russian vehicles go boom, the size of the explosion of the BTR was shockingly large.

Steve

[The_Capt]

1 hour ago, akd said:

More Avdiivka insanity (Oct. 10):

 

Mass.is.broken.

[Battlefront.com]

26 minutes ago, chrisl said:

Those older systems are really automation, rather than autonomy.

Thank you for summing things up in this way.  I think the key difference between automation and autonomy is how much the machine can do on its own to achieve an explicit goal set by a Human.  It is the difference between a machine that makes bread to a machine that can determine when it should.  Both are automating things Humans normally do, but the former has no decision making capability while the latter does.  Even if it's not much, it might be enough.

Back to the example I gave.  For sure an old school Tomahawk cruise missile can not select its own target, but it can independently navigate to a specified target without any external aids (GPS didn't exist back then).  Instead, it used internal systems to evaluate a variety of variables (including terrain signatures) in order to inform the auto-pilot.  It was even capable of aborting if it was unsure if it was where it was supposed to be.  Again, zero input from anything external.  That is autonomy, even if quite basic by today's standards.

My point is autonomous ≠ self thinking.  You can have a fire and forget drone that is specifically told to kill something that matches a specific signature within a specific geographical box.  That drone can have near full autonomy to navigate its way to the target area based on unexpected situations, such as evasive maneuvers by enemy radar.  Or it could be told to fly through a forest even though none of the trees were mapped ahead of time.  And of course you can't have true swarms without ceding full navigational autonomy.  All of this is possible without getting into the really dangerous parts of "fly over this area and kill anything that moves that is military in nature".

In short, the big challenge for drones in the near term is being cut off from communications, be it navigational or Human control.  We have the technology to solve for this already in our hands.  It's pretty easy, in fact and should be done right now.  No thinking about it, just do it.

What requires more thought is what to do about delivering harm (I'm going to include things like EMP blasts here) autonomously.  The more we try to automate that, the more information the platform needs ahead of time to make the right decision.  This necessarily limits, or should limit, how much harm it is able to do if it gets things wrong.

There's a big difference between telling a drone to go into an urban area and smash into the window of a particular building and telling a drone to go into an urban area and "kill all the bad guys".  We already do the former therefore we don't run any new risks with that.  The risk is with the latter and therefore tread lightly!

Steve

[cyrano01]

49 minutes ago, kimbosbread said:

It’s all about the economics. If 10 soliders cost as much as one tank, but each one has similar destructive capability by themselves relative to the tank, then things are ok.

 

An enhanced  soldier with as much destructive power as a tank would be impressive but surely they really have to be cost effective against the UAVs that might kill them (see UAVs chasing Russian infantry passim). Unless our 10 Starship Troopers come in cheaper than the number of UAVs needed to defeat them then these are still losing margins.

As a latter day 'Arithmetic on the Frontier' might have put it.

"Strike hard who cares—shoot straight who can— The odds are on the cheaper man drone."

 

[cyrano01]

1 hour ago, akd said:

More Avdiivka insanity (Oct. 10):

 

'This operation would make the charge of the light brigade look like a sensible military exercise.'