Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Anyone work this technology and can talk about it (without breaking whatever NDA of course)? I'm a former infantryman and ... let's just say I'm _extremely_ skeptical of this kind of technology for all but the most niche use-cases (think, AR for the driver in the hole of the truck). Obviously for pilots stuff similar to this has been around for a while, but the article is not really clear about how exactly this would be used.


I'm inclined to agree. While it might be nice in theory to have your battlefield painted with where the friendlies are or something, there are so many issues with that in practice I can't imagine it actually getting used by infantry.

Just off the top: All the electronics shit is gonna add weight, bulk, and power requirements.

It's gonna get dirty, fogs up, w/e.

It's gonna not work for one reason or another b/c it can't get a link, or it gets bashed against a humvee/wall. It doesn't get updated information, etc.

If the enemy picks it up it either shows them where to shoot your guys, or it involves some authentication system that will make it useless to guys in a gunfight anyway. (What happens if your guys accidentally switch headsets after lunch?)

It adds a bunch of extraneous bullshit in the form of information you don't need to be dealing with. Ever get lost while listening to the radio? Now raise your hand if the first thing you did was turn the music down. When shit is going south you need to REDUCE cognitive load. Pasting extra information into someone's field of view is likely to be unhelpful.


AIUR - Augmented Infantry Universal Reality headset Patch 1.76 notes

* Fixed a bug that occasionally caused friendly units to display as hostile ones.

* Corrected AR ammo counter to correctly match actual rounds in small arm.

* Adjusted networking code to ensure headsets remain synced with team leaders.

And hundreds of other potentially dangerous and hilarious patch notes...

Looking forward to everyone saying, "My life for AIUR!" while using these.


FTFY: Corrected off-by-one error in the AR ammo counter to correctly match actual rounds in small arm.

;)


New bug: Counter follows strict definition of small arm, only displays count of rounds in the chamber while ignoring the magazine/belt.


Note: Counter underflows with weapons exceeding programmed capacity. Firing negative rounds may have unpredictable behavior.


I agree with all of this.

I'll add one more: so much of what platoon leaders do now is teach about technology. There's only so many hours in the day for training, and (even back in 2013!) the load was starting to cause deficiency in basic combat tasks. We just didn't have the time between all the requirements to get everyone proficient in the myriad of sensor tech we ended up carrying. Plus, if it's possible to break something believe me a Private will figure out how to do it.

>What happens if your guys accidentally switch headsets after lunch?

This right here is a question I hope they're thinking about. In an actual fight, you may need to pick up gear off the ground. What happens if you can't, because "authentication"?


> What happens if you can't, because "authentication"?

Aren't your radios already using cryptographic fills? What do you think the difference is?


I carried around the device that filled crypto for radios.

Most people don't know how to use them and don't understand how to do a basic function check with them. It takes someone who has taken a class in receiving data with them and probably a patient Gunny who has been doing this stuff way too long.


It totally depends on how authentication would be handled for the headset device. If it's handled like it is for radios, where fills are added on a regular cycle and anyone can pick up and use the device, it's probably fine. If it's made unique-soldier-dependent, I don't think it's a good idea.


> If it's made unique-soldier-dependent, I don't think it's a good idea.

Lol well why would we do that? Just inventing random problems at this point.


It isn't a "random problem".

It's a legitimate question about the design principles that are being used here. Is the headset "dumb" and just shows the same HUD view for everyone, or is it custom to the soldier's position in the formation? We don't know, but it's a question to think about that has actual implications for how something like this could be used or even implemented.


God forbid anyone would imagine how this would this work. I wonder which scenario is more plausible, the one you responded to, or the one linked below?

https://news.ycombinator.com/item?id=26652386


Well let's get it into troops' hands and give it a go and find out!


I take your point, but I think you're unreasonably optimistic. It's worth a try, but $22 billion?

On the other hand, maybe you're right. We're both British. What do we care if the US government spunks all this money up the wall? The only downside is if they sell it to the MOD before it's working.


> myriad of sensor tech we ended up carrying

At that time, did you feel like that tech was useful?


No, and after a month or so I told everyone to drop it and put it back in the storage container where it sat for the rest of the deployment. There were some exceptions to this (some IED-defeat devices, for example).


I honestly don’t see this being used by infantry for all the reasons you list foremost being that by the time the MIL STD folks are done with it, it’ll be way too heavy.

Most likely HQ so the brass can play command and conquer with the infantry and at a push on vehicles similar to the Helmets on the F-32.


Do you mean F-35?


Yes, oops.


HL2 uses eye tracking for authentication so that one should be partially solved


From what I've seen the two key use cases are 1) being able to see a "gun's eye" view, enabling you to poke your weapon out of cover and fire without exposing yourself 2) faster friend-foe identification through the AR overlay. Both of those seem like significant upgrades and arguably worth pursuing, but you can be sure a bunch of other nonsense will probably get tacked on that may make it worse than useless (like giving it excessive network connectivity: it wouldn't be fun to get a forced auto-update in combat).


Actually, Microsoft's involvement came about when leadership decided that the investment in Clippy needed to be recouped. After some creative brainstorming, it was determined that soldiers needed a heads-up display to track their ammo reserves. Ammo is stored in clips, which is a natural fit for Clippy to keep track of.


Regrettably, the joke doesn't work since ammunition is kept in magazines or belts, not clips.

(Is there even a firearm commonly used by a modern military today that still uses clips?)


Those are both great concepts similar to the HUD on a fighter jet but having worn a HoloLens before they'd have to make significant hardware enhancements for it to be remotely practical to wear in combat. They're chunky, have poor field of view, poor brightness and contrast, require substantial power and compute capacity to be towed along.


What compute capacity? The compute device is in the headset, it's a self-contained unit and runs off of a 15W USB charger, with 2 hours of internal battery.


The PVS-21 is technically an AR night vision system, especially with the optional HUD video module so you already have battle tested AR systems used by infantrymen.

For tanks etc. AR is very useful and is already implemented at least outside of the US, Elbit has a version of their F-35 helmet for tanks and armored vehicles and the Israelis seem to be happy with the situational awareness they gain.


Right, but PVS-21s aren't networked devices are they? I thought the AR "features" were pretty limited and was more an evolution that combined what previous NVG and thermal imaging had done as separate devices. I haven't used them before.


Well they can receive and display a video input from an external module some of those modules are networked and provide sensor fusion.

I’ve seen modules that stream live navigation data, blue force tracking and video feeds. The only difference is that it’s a bulky external module which connected via a cable.

You also have small clip on modules for things like thermal imaging and video recording/streaming.

Other than that the PVS-21 is an AR headset the intensified image is projected on the lenses which allows the user to see through the headset just like any AR headset and it can display data from an external source.

https://www.steiner-defense.com/imaging-systems/cehud-confor...

Edit: It’s just not as sexy and tech loaded as the hololense but honestly I think the PVS-21 model is better the interface with the HMD is optical and the modules become smaller and more capable as technology advances.

It looks like what used to take an external box the size of a hand radio is now wireless has an OLED display and fits in the screw cap of the optical interface port.


From the perspective of infantry, the end goal would look like shared information: if any one else (other infantry, air support, satellites, drones) sees an enemy, you'd be able to see them as well, even through foliage, walls, etc. Being able to ping a location for your squad to see, marking targets for air support. I think there are a lot of practical applications to counteract fog of war.


Imagine being given a mission and being able to recreate the target area in VR and walk around while you conduct your estimate, orders, and rehearsals.

Or imagine being able to look over a piece of ground and see the location of your people annotated through AR. Or driving a route and seeing your planned route annotated.

Yes please absolutely get this into my hands.


Ok, now imagine you have to hump an extra 10lbs of crap in 120 degree heat for weeks on end while the glasses are fogged up and you got a sunscreen smudge on the lens. And your field of view is now greatly complicated with extraneous information you don't need to know like ammo capacity of your current magazine. And you need to low crawl through a drainage/sewer ditch and through a hole in a fence to get to another position with all that crap on your head/body without getting tangled up.

And now your patrol got delayed for some reason so now none of that crap is charged or working anyway. And your glasses or the controller got dislodged during a sprint to cover. Or you need to find the reset button for some reason.

You can have my set. It might be a valuable training tool though.


> You can have my set.

Thanks, I'll take it!

You can say the same as you have about any technology. Why lug around a big heavy rifle when you could carry sharpened sticks? I guarantee you people said exactly the same about the first radios, for example. Now you wouldn't even consider leaving the wire without a radio under any circumstances whatsoever.


>You can say the same as you did about any technology. Why lug around a big heavy rifle when you could carry sharpened sticks?

Because the enemy doesn't use sticks, first of all.

>I guarantee you people said exactly the same about the first radios, for example.

They did not, because it represented a fundamental change in how ground warfare could be conducted and not everyone carried them (not everyone carries them today either, which is telling). Namely, coordinating accurate indirect fire.

>Now you wouldn't even consider leaving the wire without a radio under any circumstances whatsoever.

Because I want to be able to call in air-support and indirect. The difference with all these examples is that the technologies you are mentioning represented a game-changing way in how to enable infantry-support operations. It's not immediately clear to me that the same is true for a AR/VR helmet system used by infantry. It's possible that other branches could find uses for it in the same vein. Like the FSO using it to "see" the FLT and better coordinate indirect fire.


> I guarantee you people said exactly the same about the first radios, for example

This doesn't prove anything and is 300% survivorship bias. There are thousands of tech innovations that were utter shit and we never hear about them. AR on the battlefield as describe above is a COD player wet dream and has no basis in reality.

https://www.popularmechanics.com/military/research/a25644619...

> King estimates that the average soldier goes into action with a hefty 20 lbs of batteries.


> AR on the battlefield as describe above is a COD player wet dream and has no basis in reality.

I can only offer that I want to try it.


The map is not the terrain. A VR walk through isn't as valuable if the layout of the target area changes. Same with troop locations and routes. Any training or planning use will still come down to quality of intelligence.

In general, it seems very unlikely that all of this will work coordinated in battlefield conditions. It would be like pulling off an AR MMORPG, except everyone is carrying their Xbox on their back and running around randomly


> The map is not the terrain.

Where did I say it was?

Let the recce give me a 3d sketch map of the target area as they understand it and let me walk around in it. Better than them trying to describe it and me imagining it!


> I'm a former infantryman and ... let's just say I'm _extremely_ skeptical of this kind of technology

The question is how integral can you make this, without retraining everyone for it.

For this to be successful there needs to be a whole generation who is probably 16 or 14 right now, who are used to putting on these headsets and be habituated to moving around in a VR world - wait till they turn 18 and put them on the battlefield at 19.

I'd imagine this is already present in CAG or some other secret division, but we've always seen how it "should work" for friend-or-foe in the Terminator HUD clones[1].

[1] - https://blogs.windows.com/windowsdeveloper/2017/03/06/buildi...


Well, it's true to a limited (but often exaggerated) extent that tech used in JSOC or other Tier 1 units eventually filters down to the grunts, but the limiting factor here is that the folks in CAG etc are operating at cognitive abilities way above the standard infantryman (sorry, it's true). So things that are interesting tech (NVGs) that are easy to use, yeah they will filter down. Suppressors being standard for USMC infantry units is another example. "Push button, see at night" and "screw on, shoot gun more quieter" vs "N hours of training and re-learning your field craft to distribute the weight".


I work on AR/VR in the pharma field. It's not super close to downrange applications, but I may be able to handle some questions.

To your concerns, I think they are very valid. Weight, as always, is a primary concern for the infantry. The Hololens2 is not a light system to just wear, let alone with a helmet. Additional issues like battery, signal, and other unforeseens are likely large concerns too.

That said, I think that the ability to quickly and seamlessly integrate all the data that is coming in may have some advantages. Giving warfighters a way to view all the data in real-time is a goal worth spending some money on. The recent Nagorno-Karabakh war has shown that the digital/cyber aspect of war has a likelihood of being a vital factor in future conflicts.

https://en.wikipedia.org/wiki/2020_Nagorno-Karabakh_conflict


Integrate data for who? I feel the common footsoldier is not the aim here, this is just more metrics for the comms to use and if that.

My scepticism lies in the fact that bulletproof vests, widely used across different countries, was not an invention of the overfunded military-industrial complex, but an ex marine who hacked it together in the back of his garage with a rusty fork.


That's not true. Modern body armor was available in Vietnam [1], and was the result of an ARPA research program. It was expensive and rarely issued though until later, but the technology (boron-carbide plates) is similar to what is used today. You're thinking of Richard Davis, the inventor of the police-issued Kevlar vest. He invented it in the mid 1970s, after getting in a gunfight while working as a pizza delivery driver.

[1] https://docs.google.com/document/d/1iw3_4RGaDlSDXmeYkSYu-rQu...


Former Infantry here also but still serving part time.

I think the immediate use is for training. The Army is big on simulators for all MOS's because it saves resources, reduces training complexity, and reduces risk. Sims such as VBS3, EST, VCOT, UGT are mostly required these days for various purposes. This could even possibly function as a better MILES or replace the fleet of OPFOR vehicle analogues at NTC. So, potentially there is money to be saved while creating more realistic training.

Maybe in a few decades the software stack will be solid enough and the hardware miniaturized enough to use in combat. The similar success story might be the FA's venerable AFATDS, which has been around for decades and very reliably brings rounds on target.

Worst case, it ends up as extra weight for graded positions to carry in RS. ;)


Opinion. The goal is to make certain people a lot of money. See F22 for example.

If actually works that’s a bonus.


Imagine a rugged version of Google Glass that can act as a HUD with maps and current location and enhanced situational awareness.

I think the long term goal is to have something like the XCOM 2 Specialist class.


I don't work with the tech, but is there a use case for operating drones in the field?

You want to peek around a corner or under a door, drop a tiny robot. Operate it with a handheld joystick and get a camera feed into your headset while still maintaining your regular field of vision.

Final product won't necessarily take the form of the hololens. Might just be an attachment to whatever standard equipment they have, more like a Google glass.


Not really possible with HoloLens. This is AR not VR or a HUD.


I'm curious, what's the obstacle? Can't they take the display from the hololens, make it smaller and integrate it into one eye of an existing headset that the military uses?


Existing HoloLens devices show semi transparent holograms overlaid on the real world so to speak, in a viewable frame size resembling something like a oversized postage stamp. (That is, holographic content does not fill your entire field of view.) It's not conducive to showing 100% opaque video content for a live drone feed, in my opinion, due to technical limitations (opacity, rendering proximity) and human comfort issues.

I appreciate that they're offering "custom" units and could theoretically fix these issues. But it _sounds_ like it'll be more of a ruggidization and compliance realignment of existing hardware. Pure speculation of course.


HoloLens 2 devices offer an expanded field of view of approx 60 degrees. It's still semi-transparent, and opaque content has improved but still can't outshine looking directly at lighting.


A video feed has areas of lighter stuff along with areas of darker stuff. (Most) AR systems work by projecting lasers on to a glass plane in front of you, adding light to your FOV. How do you project darkness on a transparent background? The video feed won't look that great in AR.


Not true, you can watch YouTube in HL2 out of the box. There are videos of hobbyists visualizing a self-driving RC car from the HL2.


Mission planning in the field, reducing costs for the field tents that the more senior officers use, 3d maps with plans shared to the team

Imagine Battlefield like HUD for the soldiers, green is friendly, red is adversary. Think of these as a part in an overall sensor fusion and information asymmetry in the field. There are larger initiatives around info fusion across the branches and the real-time access to those who need to know


I think in these cases the actual utility of the product does not matter. What matters is that someone gets a government contract and something exchanges hands.

Something can be a failed national health platform, or toys like this one.


Infantry here too, I think it would be cool for the TC to have the blue force tracker, maps, and any other sensors available in a heads up display.

Also, seems like a no-brainer for tankers.

I do not want to carry this shit on a patrol though.


Agree, the Armor community might really like this stuff. Especially if it can be paired with a sensor suite on the outside of the vehicle, so the TC (or maybe everyone) in the vehicle can "see" through the hull. The complexity of determining which surfaces within the truck should show as translucent in the HUD seems like a pretty tough problem to me though.


HL2 consumer version weighs less than 1lb and will likely be built directly into your helmet. Might get to 1lb for extended battery life


1lb is still a lot on your head considering everything else you have to hang off of it.


Training - simulations.

Eventually, a Danger Room anywhere.


they're running Windows 7 because Candy Crush isn't preinstalled


Is used for me in 10 years when it all trickles down. I can’t wait.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: