I read the article, but I couldn't find the part where they unveiled the secret to quieter small drones.
It looks like they're improving the fidelity of computational models for acoustics, which is an important step.
But that doesn't imply that any change to the design of the drones has actually been identified yet, even if that's what these models will be used to do.
I would be shocked if their R&D isn't far ahead of what they share publicly. We've seen in the recent conflict between Azerbaijan and Armenia that drones have become a major factor in modern battlefields. And not just surveillance, as the article talks about, but essentially as affordable missiles.
At least there's an obvious civilian benefit from this work.
I wouldn't be. AFAICT all the classified stuff in most defense systems is usually applications of existing technology. Classified R&D is large-D, small-R. They only get significantly ahead of the industry when nobody else is doing research. Such as when the only applications or scientific value is in warfighting.
Sometimes "nobody is commercializing the research" can look a lot like "nobody is doing research." In holography, for example—no civilian market demand, so even in academia, the researchers tend to retread the same ground, "forgetting" all the existing research, because there are no productized civilian applications around to remind them that X/Y/Z can already be assumed.
It’s less that they’re explicitly doing the same thing while not citing existing research; and more that they’re constantly inventing new and different ways to do foundational things, rather than taking some of the things that have already been proven to work and “moving up the tech tree” to doing more interesting things.
(Imagine if you will: everyone constantly doing research into new lightbulb technologies — so we’d get the LED a century sooner! — but nobody ever researching the implications of evacuated incandescent bulbs [vacuum tubes! CRTs!], nor into the implications of miniaturizing/printing LEDs, etc.)
It’s kind of like where fusion research and ML research were during their “dark decades” — everybody doing breadth-first searches of the possibilities, nobody digging much deeper into the implications of research that showed promise.
In all three cases, it’s because it’s hard to model those implications without a real experimental prototype; and because, in all three cases, costs to do real experiments (for reactors; for training datasets; etc.) shoot up exponentially as you “go deeper” — as you need to do those experiments in the context of a working example of everything you know so far.
The true issue is one of team size dynamics. Particle physics has figured this out: to continue foundational research, you need a few large teams with huge grants/budgets (to build particle accelerators), rather than many small teams (that can’t individually afford to build particle accelerators.)
But other fields haven’t managed to have the same epiphany, and so in those fields, small teams just continue to explore the parts of knowledge-space they can “reach” with a small team — which isn’t really too productive.
(ML got going again because large companies figured out basic research into ML could be a loss-leader for renting out time on specialized ML-accelerator hardware. Nuclear science never slowed down, because governments were doing the basic research with huge teams+budgets from the start. Etc.)
Materials research is an exception to your claim. Aviation and weapons research depend heavily on exotic materials that are almost certainly NOT found in civilian applications. E.g. nuclear weaponry, stealth coatings, metamaterials cloaking devices, camera systems, hypersonics, directed energy weaponry, plasma generators, etc.
I'd believe that it is, but why would they publicize that they've unlocked "the secret to quieter drones" if that were a classified development?
Even if they can't say that "the secret is to change the fans like so", they'd at least say "thanks to this work, the latest drone prototypes are now half as loud as the old ones" or something to indicate the progress.
If the fact that a noise reduction was achieved was itself classified, then the headline would trumpet an advance in computer modelling, which is what the article actually talks about.
> I'd believe that it is, but why would they publicize that they've unlocked "the secret to quieter drones" if that were a classified development?
Marketing. For a potential adversary to gain useful knowledge from this article, they'd need some numbers to understand the extent of the capability. They're not learning much from "we're using computers to make shit better", they already know that.
It is not uncommon for the military to publicly announce something, yet keep the specs a closely guarded secret.
One way is to have balanced yet asymmetrical blades. Instead of all blades reinforcing each other with the same sound, you get different blades each making sound on a different set of frequencies. You get less sound over a wider spectrum.
A picture of a propeller craft while talking about rotors, check.
I can't imagine it'd be much more than tweaking helicopter designs to suit miniature versions. Designs they've had for decades. Heck, I wouldn't be surprised if they already have rotor drones using what they're talking about.
The article specifically discusses rotorcraft, but with a photo of a prop-driven airplane. Rotorcraft generate their lift using rotors. Prop-driven airplanes are not rotorcraft - they generate their lift with wings (and the rotor provides propulsion only).
The article mentions rotorcraft, but I would imagine the technology works for propellers just as well, wouldn't it? And many drones in the army are exactly as pictured, propeller driven fixed wing aircraft.
I agree it does seem to be confusing as to exactly what this applies to though. The wording says one thing, but the picture and common sense seem to indicate it may not be as specific as that? /shrug
I've essentially done this. I was in a location where flying the drone was 100% going to get me in trouble. So, I fired up the app, started the camera, and carried the drone above my head as I ran down the street. The gimbal was active and in the end, looked like a drone shot. Did the same thing shooting from the backseat of a convertible. People still ask how I got the footage.
Unfortunately, I don't. It was all for a corporate gig, so the footage was theirs in the end. The link I had to the video is no longer active. They used it, and the got rid of it I guess.
> Everyone who flies drones knows they're annoyingly loud
I believe that a lot of that comes down to the current state of affairs simply being the cheapest or most cost effective solution.
I don't know what the military is currently investigating, but if memory serves me well then the commercial industry has been long working on more silent designs (in particular regarding propellers). Of course, even there it is a constant trade-off between effectiveness (regarding propulsion) and what is considered acceptable discomfort for passengers or the environment (regarding audio).
I believe that drones could already be made way more quiet with long existing technology. The key question here might be: how much shorter do you want your flight last, and how much more expensive do you want your drone to be?
“Boundary layer theory” was the tongue-and-cheek way my undergraduate professor would hand-waivingly dismiss hard empirical problems that couldn’t be rectified with theory
There's been work in that area. Uneven blade spacing helps.[1] One of the US "stealth helicopters" sounds more like wind noise than the distinctive "thud thud thud" of most helicopters, according to one of those books by ex-special forces types.
Drones might benefit from that trick. But you pay a penalty in performance and energy consumption.
Well, you could make boutique drones exploiting state of the art tech that are so expensive you can only afford three of them. Or you could just buy load of cheap throwaway drones and swarm your targets with them.
Remember, lots of drones used as ISR assets had unencrypted video downlinks because the chance that the enemy could also see the downlinks was considered less bad than the cost of rolling out state of the art crypto across coalition forces
Directional radio antennas tend to be rather large, and I'm not sure it's even feasible to get rid of side-lobes to the point where they're undetectable (or at least, with a signal too weak to interpret).
Directional optical links might be feasible, but probably a lot more expensive than a simple radio.
> The goal of her work is to develop a method of obtaining the parameters of the airfoil boundary layer from an existing high-fidelity computational fluid dynamics code, without requiring any more effort from the code’s end-user than was required previously, she said.
So the goal is to make this a more generic computer model to avoid the cost of wind-tunnel testing using a fluid dynamics model. I don’t know from the article if any of this has actually unveiled anything - sounds like the goal of the work, which is progressing Ok.
From 2013 onwards, all (non-classified) work funded by the US government must end up in an open-access repository, either immediately or after a 6 or 12 month embargo period. Here's the policy memo: https://www.science.gov/docs/ostp_public_access_memo_2013.pd...
I wish more science reporting was written this way. Take the big topic apart into a finite number of blocks and work through them: introduce concept, explain concept, impact concept, tie to next paragraph. In the end I was able to follow through multiple topics I knew noting about and get a high level understanding.
What I find from media science reporters is often missing background or context (or just flat out wrong) with oodles of impact (supposedly). From scientific journal articles I often lack the required background to even follow the abstract and almost certainly not the jargon filled paper (whether they are written that way on purpose I have my own opinions about).
I know you can't spend time explaining every building block concept, but the selection of what gets background and the succicintess of the explanations here is really nice for me.
I wholeheartedly agree, you a articulated what I felt when reading this overview which is why I submitted it. Much respect to the writer and their organization for having high writing standards and making it accessible to curious outsiders.
Offshoot topic but why on earth didn’t mosquitos evolve to fly quietly? They could bite me all day long and I’d never know but I can hear them almost across the room.
And sometimes they seem to be quiet and other times it’s a loud buzz. Really seems like they’re not optimized for their job?
Ticks on the other hand are silent, migrate to hard to locate parts of your body, you can’t feel their bite and are very crush resistant. When can we exterminate them all?
I think it's because they evolved to bite humans and animals sleeping outdoors. The night is full of sounds of crickets, frogs, and other creatures that can drown out a mosquito's hum.
Quiet bug-proof houses are a new invention. Very few mosquitoes successfully make it inside, bite, escape the building, and find a suitable place to lay their eggs. If most mosquitoes get killed after entering the building, natural selection may be pushing for mosquitoes that don't enter homes at all.
Noise canceling works on the principle of destructive interference. Why can't drones add noise generating components which are just out of phase with the primary noise generating components? I assume this would lead to a loss of efficiency, but perhaps not enough to outweigh the benefits?
52 comments
[ 2447 ms ] story [ 2952 ms ] threadIt looks like they're improving the fidelity of computational models for acoustics, which is an important step.
But that doesn't imply that any change to the design of the drones has actually been identified yet, even if that's what these models will be used to do.
At least there's an obvious civilian benefit from this work.
[1] https://en.m.wikipedia.org/wiki/Dual_EC_DRBG
(Imagine if you will: everyone constantly doing research into new lightbulb technologies — so we’d get the LED a century sooner! — but nobody ever researching the implications of evacuated incandescent bulbs [vacuum tubes! CRTs!], nor into the implications of miniaturizing/printing LEDs, etc.)
It’s kind of like where fusion research and ML research were during their “dark decades” — everybody doing breadth-first searches of the possibilities, nobody digging much deeper into the implications of research that showed promise.
In all three cases, it’s because it’s hard to model those implications without a real experimental prototype; and because, in all three cases, costs to do real experiments (for reactors; for training datasets; etc.) shoot up exponentially as you “go deeper” — as you need to do those experiments in the context of a working example of everything you know so far.
The true issue is one of team size dynamics. Particle physics has figured this out: to continue foundational research, you need a few large teams with huge grants/budgets (to build particle accelerators), rather than many small teams (that can’t individually afford to build particle accelerators.)
But other fields haven’t managed to have the same epiphany, and so in those fields, small teams just continue to explore the parts of knowledge-space they can “reach” with a small team — which isn’t really too productive.
(ML got going again because large companies figured out basic research into ML could be a loss-leader for renting out time on specialized ML-accelerator hardware. Nuclear science never slowed down, because governments were doing the basic research with huge teams+budgets from the start. Etc.)
My favorite was when they showed the image with the working microscope.
Me> They only get significantly ahead of the industry when nobody else is doing research.
Even if they can't say that "the secret is to change the fans like so", they'd at least say "thanks to this work, the latest drone prototypes are now half as loud as the old ones" or something to indicate the progress.
If the fact that a noise reduction was achieved was itself classified, then the headline would trumpet an advance in computer modelling, which is what the article actually talks about.
Marketing. For a potential adversary to gain useful knowledge from this article, they'd need some numbers to understand the extent of the capability. They're not learning much from "we're using computers to make shit better", they already know that.
It is not uncommon for the military to publicly announce something, yet keep the specs a closely guarded secret.
EDIT: https://en.wikipedia.org/wiki/Fenestron
I can't imagine it'd be much more than tweaking helicopter designs to suit miniature versions. Designs they've had for decades. Heck, I wouldn't be surprised if they already have rotor drones using what they're talking about.
Propellers are rotors. Rotor is the generic term for something that rotates, while a propeller implies movement along the axis it is providing thrust.
In other words, a propeller is a specific type of rotor.
I agree it does seem to be confusing as to exactly what this applies to though. The wording says one thing, but the picture and common sense seem to indicate it may not be as specific as that? /shrug
1: https://www.avinc.com/tuas/puma-ae
Everyone who flies drones knows they're annoyingly loud.
They have been getting somewhat better though, and you get what you pay for.
I believe that a lot of that comes down to the current state of affairs simply being the cheapest or most cost effective solution.
I don't know what the military is currently investigating, but if memory serves me well then the commercial industry has been long working on more silent designs (in particular regarding propellers). Of course, even there it is a constant trade-off between effectiveness (regarding propulsion) and what is considered acceptable discomfort for passengers or the environment (regarding audio).
I believe that drones could already be made way more quiet with long existing technology. The key question here might be: how much shorter do you want your flight last, and how much more expensive do you want your drone to be?
Not gliders...
Drones might benefit from that trick. But you pay a penalty in performance and energy consumption.
[1] https://www.globalsecurity.org/military/systems/aircraft/mh-...
Well, you could make boutique drones exploiting state of the art tech that are so expensive you can only afford three of them. Or you could just buy load of cheap throwaway drones and swarm your targets with them.
Remember, lots of drones used as ISR assets had unencrypted video downlinks because the chance that the enemy could also see the downlinks was considered less bad than the cost of rolling out state of the art crypto across coalition forces
You would think they would have highly directional signals by now.
Directional optical links might be feasible, but probably a lot more expensive than a simple radio.
So the goal is to make this a more generic computer model to avoid the cost of wind-tunnel testing using a fluid dynamics model. I don’t know from the article if any of this has actually unveiled anything - sounds like the goal of the work, which is progressing Ok.
EDIT: Link to the paper. Yours for $30. Ugh. Why is this costing something when I already paid for the research? https://vtol.org/store/product/computation-and-extraction-of...
For this, I'd look in DTIC, though I couldn't find it yet: https://publicaccess.dtic.mil/padf_public/#/home
404 for me. Not sure whether that's because I'm not in the US or because I'm using a mobile browser or whether you messed up the link.
What I find from media science reporters is often missing background or context (or just flat out wrong) with oodles of impact (supposedly). From scientific journal articles I often lack the required background to even follow the abstract and almost certainly not the jargon filled paper (whether they are written that way on purpose I have my own opinions about).
I know you can't spend time explaining every building block concept, but the selection of what gets background and the succicintess of the explanations here is really nice for me.
And sometimes they seem to be quiet and other times it’s a loud buzz. Really seems like they’re not optimized for their job?
Quiet bug-proof houses are a new invention. Very few mosquitoes successfully make it inside, bite, escape the building, and find a suitable place to lay their eggs. If most mosquitoes get killed after entering the building, natural selection may be pushing for mosquitoes that don't enter homes at all.