I'm lost in the weeds, trying to figure out what is _actually_ going into the army's replacement for MILES. So far, I've got "FRICK lasers" and "LUVS ME ar/vr, simple as" and that's about it. Does anyone have good info on this? I'll include my notes so far.
SYNTHETIC TRAINING ENVIRONMENT (STE)
This is the US Army program for JSE.
2024: Official pages are stubs with nothing on them.
https://asc.army.mil/web/portfolio-item/synthetic-training-environment-ste/
https://www.peostri.army.mil/synthetic-training-environment-ste/
2023: STE is NOT LASERS, because LASERS is not FUTURE.
https://archive.md/d5YZV
>“We can arguably go out and buy a new laser-based system that will be a new version of the old technology. But our senior leaders have told us not to do that: don’t go out to buy new versions of old technology,” said Brig. Gen. William Glaser, director of the Army STE Cross-Functional Team[.] Ray Chandler, Lockheed Martin synthetic training environment campaign manager and former sergeant major of the Army, told reporters at one of the company’s Orlando facilities that the current version of MILES only supports about 45 percent of a brigade combat team that cycles through a combat training center. “So, 55 percent of the combat power and brigade does not have the ability to play in the battlespace for the training part. That’s huge. If you can go from 55 to 100 percent of all those enablers and allow everyone to be a part of the game, it’s an exciting thing,” he said.
2019: This has involved prototypes of a system called SIMRES.
[ibid.]
2021: The program has been part of a holistic force-on-force update.
https://archive.md/Oup3F
>"A lot of times, people will think we are trying to replace our live training environment with virtuals," said Maj. Gen. Maria Gervais, director of the STE Cross Functional Team. While STE is a priority, Army senior leaders have directed the training simulation community to develop more effective sims for direct-fire, force-on-force training, she said. "We have been directed to focus on our direct fire, so ... we are going to replace our MILES capability," Gervais said. "We are also going to focus on [simulated] indirect fire, and we are going to focus on counter-defilade fire"[.]
STE, continued:
2020: One component of STE is eBullet, aka WULF.
https://archive.md/epMrF
>“MILES has not been able to replicate or simulate many weapons systems,” said Frank Tucker, live-training technology lead for the STE team at Army Futures Command. “It’s only been able to simulate 60 percent of the current inventory of weapons.” Worse, he said, it can’t simulate enemy weapons or potential future weapons at all. But with eBullet, Tucker told me in an interview, adding a new weapon could be as simple as tweaking a few factors in the model to create a “digital twin” of an existing or potential technology. “The Army now can dream up a new weapons system and synthetically deploy it in a training exercise to see how it works or doesn’t work, without bending metal,” he said. “It can just simply be a model that’s embedded in e-bullet.”
2020: eBullet/WULF can incorporate CBRN and EW effects.
[ibid.]
>For more complex phenomena, like electronic or biochemical warfare, the service could just take existing Army effects-propagation models and plug them into the eBullet suite. “You can do some very, very interesting things we never had in live training,” Tucker said. “We’ve never been able to train this stuff — never.”
2020: eBullet/WULF can play nice with IVAS.
[ibid.]
>[Y]ou could superimpose virtual dust plumes, tracer rounds, and explosions on the wearer’s field of view, giving instant and realistic feedback. [...] It calculates precisely where each shot will go and transmits that data to the soldier’s augmented-reality goggles (either the new ENVG-B or the forthcoming IVAS)[.]
2020: eBullet/WULF's first mature demo (TRL6) is mortar fire correction.
[ibid.]
>[M]ortar crews need to carry a 40-pound, $60,000 gyrocompass called a Talon II to ensure everything’s correctly aligned. [ebullet/WULF is] twice as good as the existing Talon II, Tucker said, with errors under one milliradian: That’s less than one-seventeeth (0.58) of one degree.
STE, continued:
2020, 2023: This depends on ONE WORLD TERRAIN [OWT], a Vricon/Maxar product.
https://archive.md/WsK2X
2023: The infra that will integrate the mapping with the AR/VR environment is called the STE Information Suite.
https://archive.md/KaA7y
2023: There are 26 people on the STE team.
[ibid.]
2023: The STE component for shooty drills is called the Squad Immersive Virtual Trainer (SiVT).
[ibid.]
COMMON SIMULATION ENVIRONMENT (CSE)
AKA JOINT SIMULATION ENVIRONMENT (JSE)
AKA JOINT TRAINING SYNTHETIC ENVIRONMENT (JTSE)
This is the joint USAF and Army program.
2023: Seemingly exists so that flight simulators can pair with NOTMILES.
https://archive.md/qvVDf
https://archive.md/Q1FMz
https://dvidshub.net/news/451062/
https://dvidshub.net/image/7962476/
2023: Issues with F-35, delays, drama.
https://archive.md/kX72k
2024: Contracts have been awarded.
https://archive.md/wLTHp
2024: The shared equipment model library is called the Next Generation Threat System (NGTS)
https://archive.md/Q1FMz
That's all I've got.
Seems like the gameplan is to have ground troops wearing AR goggles that you can project all the heavy assets onto, while the heavy assets are being controlled by someone back at base in a VR sim. Hit detection is going to be done by shooting a virtual bullet using the heading of the weapon with hits being detected by the target's GPS position, though there is probably also going to be some sort of LOS sanity check so that you can't just spray your M4 into a house 1km away and kill everyone inside it. This also means that you can program any weapon to have any characteristic, so if you want to see what it would be like if M4s were secretly magic that project a foot wide laser beam of death but only had a range of 20 yards, you can just set that up. Neat idea for a system, though I do think that the inability to have heavy assets do things like knock down trees and other destructive elements to reshape the battlefield is a weakness of it, but I also have no idea how you would do that safely.
Do you have a specific question?
Most basic level: What does simulated firing actually look like? Are there training / simunition rounds? Is is purely dry fire? What does simulated firing of a vehicle's cannon/gun look like?
Taking fire and modeling the game state: How is everything linked up? Lack of laser feedback implies the need for netcode, which adjudicates between P1's sensors and P2's movements and hitbox.
And so on.
I don't have one. I have lots of frickin acronyms but still no worked example for grugg brain.
>What does simulated firing of a vehicle's cannon/gun look like?
In case it isn't obvious: This also includes questions about faking data for the vehicle's own onboard optics, emergency diagnostics, and FCS, which are going to expect there to be a projectile.
>no LASERS because LASERS ARE NOT THE FUTURE
how the frick are they planning on modeling all the micro terrain in JRTC and NTC and detect the body postures of soldiers to detect hits from the E BULLET
Too long didn't Sneed give rundown
They are not getting rid of MILES
What's going to happen? For years they have been insisting on a MILES sunset of around 2027. All of this new shit is supposed to be done shortly before that in 2026.
The can will be pushed down the road until they figure out a way to replace the current functioning system of lasers+sensors. If anything, they'll integrate the acronym soup mentioned ITT into the next system with something that very much resembles MILES, because there is frick-all as an alternative available for the foreseeable future.
IMO, the dicks lugging MILES gear around are going to get more random sensors and GPS suites to tell them they're dead from shit they didn't even see. I think the biggest improvements would be artillery casualties, because the current system is kind of non-functional.
There are already several off the shelf MILES like systems that include more networking, gps, better sensors and lasers to solve all the problems they're talking about
List them, please? Trying to learn what's state of the art in force-on-force training is torture. These nu-IT niches combine the worst vices of military and tech.
How does it currently handle fires bigger than small arms? I know there are damage sensors for vehicles to let them hit mines/IEDs and the like, but I can't recall seeing them use MILES offensively.
>I think the biggest improvements would be artillery casualties, because the current system is kind of non-functional.
The new mortar thing does look promising
But I can't, for the life of me, imagine the pain of the systems' developers upon hearing the media guys just casually drop, "Oh btw, CBRN? Electronic warfare? All-Domain Ops? Yeah, those will all be pluggable modules. Buddy, look: We're MODULAR. We're AGILE. It'll be EASY."
I also find it disturbing that so much of the PR about NOTMILES consistently claims that going from sims to live-fire will be seamless. So seamless that we'll be pairing NOTMILES with IVAS and co. in combat environments!!
If everyone is already networked with the IVAS system and it's accompanying datafeeds, it's actually probably trivial to have CBRN stuff since it's just a geofence hitbox that grows or shrinks over time. EW and All Domain might be harder, but once again, if you keep it simple as a geofenced hitbox, or nested set of those of increasing strength, that degrade the targeted function, that seems pretty doable.
>just a geofence hitbox that grows or shrinks over time
You think so? It could definitely be done that way, but it feels bordering on pointless if it amounts to Fallout style rad damage.
Also, big bumperino for any and all experiences with MILES like systems.
I mean, if the point is to have the troops learn how to handle getting hit with it, I'm sure the US mil has accurate models of how it would move and can set the geofence to behave as it would IRL which would be a pretty good training mechanism if you have the troops able to "negate" the damage by donning the correct gear, which will probably require an observer or them self reporting. It also would be a good way to see train the troops to respond to a time pressure situation if they have the Redfor continue to attack them conventionally while the maps show the "poison gas" rolling closer. They also claim to have the ability to put shit onto the IVAS display, so you could even have the physiological pressure of seeing the gas cloud rolling in instead of just a red line on a map. It's not perfect, but it would certainly be good enuff for training and far better then current methods of just having an observer make the call and trying to convey the information verbally.
Where's the main locus of the computing behind this?
If you mostly crunch the dynamics of the sim on a server and push state changes out to training wearables that are basically operating as kiosk-mode devices, you're doing something tractable and interesting for training, but probably useless for combat.
If you do as much as you can client-side, you're imposing power, heat, and weight requirements to do something that you'd presumably rather have directly integrated into standard kit.
The problem of developing the training wearables parallel to everything else seems to be in tension with "everything else."
The IVAS system is already operating under the paradigm of having central command sending out data and signals to soldiers in the field and them feeding data back, so while I can't speak to how they're sending it back and forth, the Army clearly has a belief that they'll be able to conduct some form of communication under enemy surveillance and EW. As to the split, you can look at the commercial Quest systems that are capable of running a pair of high res displays off of integrated hardware at a very low price point. The majority of the bulk comes from the screens themselves, while the IVAS system doesn't have that issue and also can distribute the computing to a device that's just woven into the soldier's MOLLE webbing, though thinking for a moment, I can't remember if that was just the interface device, battery, computation hardware, or some combination thereof. You can also have the IVAS system running the power and computationally expensive stuff of rendering non-real weapon systems interacting with the battlefield during training and just not use that capability to preserve power when in "real" conditions, though even then the IVAS system inherently is going to be doing some rendering to put the information into a HUD format.
Neat. You happen to have any technical materials on the current IVAS architecture? Systems shit, horrible milspec hairball graphs, developer requirements, what have you.
What do you propose, then? There isn't anything of the sort in existence, besides your instructor looking you over and telling you a strap is out of place. WIFI enabled mopp suits that tell you that your taint is too tight? Immersive simulation of NBC shit is all going to look like fallout rad damage with maybe some prop gas and special fx to sell it when someone comes down and says "okay your unit dies in a tabun attack, lay down for a bit." You want IVAS to render clumps of AR hair falling in your peripheral? Play the Minecraft poison tick while you're all taking your simulated vx swabs? The rad bar is underwhelming, yes, but until we can make you feel like you just inhaled mustard gas without killing you... That's still as deep as it gets.
There's lots of good stuff out there for CBRN. Tactile feedback, simulant chem kits, scent libraries, what have you. For the most basic quals, tacking a plume model onto the same AR gear that handles ballistic effects might be better than nothing. Might not be. No firm opinions on it because it's a phenomenological thing. The sensory integration either comes together to sell the fx or it doesn't.
Lmao this explains why a Sandia nerd was thirsting after my resume (phd geologist specializing in how microbes are transported through eolian processes) at our career fair. I was like why the frick would you care about my DIRB research
Why would DOE care about this? Milspec anything is usually several notches below basic science r&d in caring about model fidelity.
Will the new simulation environment support rhythm games?