People have a tendency to believe what they want to believe
even in the face of proof to the contrary.
One such example is manning for unmanned systems. So many people take it as an article of faith
that unmanned systems decrease the overall manning requirement and that’s
patently false. I documented this back
in a 2018 post which quoted Air Force generals stating publicly that unmanned
systems required more manpower than manned systems.[1] Here’s a quote that was presented in that
post:
The [remotely piloted aircraft] ... requires much more architecture than, say, an F-16 squadron, Kwast said. While the ratio of people to aircraft in manned aviation is roughly 1.5 to 1, he said, it takes about 10 people to operate one UAV at any given time.[1]Now we have the Army saying the same thing.
“It’s kind of a paradox that our ‘unmanned’ formations are larger than our manned formations,’” said Maj. Gen. Michael McCurry, a veteran helicopter pilot who now heads the Army aviation “schoolhouse” at Fort Rucker, Ala. “We have Apache [attack helicopter] companies that are just over 30 people and we have Grey Eagle [drone] companies that are 135 people [or more].[2]Tell me again, now, what’s the advantage of unmanned systems and why have jumped straight into the deep end of the unmanned pool with no evidence of its combat usefulness?
 |
| How many men does it take to change an unmanned light bulb? |
https://navy-matters.blogspot.com/2018/08/unmanned-thoughts.html
https://breakingdefense.com/2023/02/unmanned-drones-take-too-many-humans-to-operate-says-top-army-aviator/?_ga=2.228372958.2132452390.1677543794-1009727458.1647467636
The military doesn't actually want to reduce manning in itself.
ReplyDeleteThey just want to have less people in combat since they get yelled at when someone is killed.
@Lonfo. Agree 100%. I think it started with the over usage of cruise missiles and higher ups military and civilian realized we can pretend to do something and it's just some money thrown into the fire when u lob a few CMs at whomever "deserves it", no US soldiers or US pilots are in danger and no one gives a sh#t if a few innocent brown bystanders die, funny to think that its the same time CMs became popular, military started calling it "collateral damage", when we all know we just wasted some civilians.
DeleteCMs and now drones are perfect for politicans: pretense of doing something at "no costs".
Why change anything? It would disrupt the unnatural flow of money within the MICC! Money in motion stays in motion, money at rest is an annomoly that must be immediately allocated to my constiuants.
ReplyDelete"Tell me again, now, what’s the advantage of unmanned systems and why have jumped straight into the deep end of the unmanned pool with no evidence of its combat usefulness?"
ReplyDeleteGray Eagle endurance - 25 hours
Apache endurance - 3 hours
That accounts for 8.33 more people right there. Once a little autonomy is introduced we will have acheived the battlefield ability similar to a wolf running down a deer. With endurance on our side so is time.
Delete"Gray Eagle"
DeleteAs the Air Force has publicly noted, large UAVs have no combat survivability and, therefore, no combat effectiveness and the AF is moving away from them.
The AF is not moving away from them. It is shifting to UAVs that are more stealthy, more attritable, or both. See Loyal Wingman, RQ-170/180.
DeleteAlso check out this Mitchell Institute (think tank) report.
https://mitchellaerospacepower.org/wp-content/uploads/2022/12/The-Next-Frontier-UAVs-for-Great-Power-Conflict-FINAL.pdf
The Grey Eagle- 4 Hell Fire missiles
DeleteThe Apache-16 Hellfire plus a 30mm with 1200 rds.
Grey eagle operates at 20,000ft in UNcontested air space.
Apache-flown in nap of the earth profile in combat to avoid radar, and actually used to take out AA radar installations during Desert Storm.
In uncontested airspace, an AC-130J has almost 4 hours loiter time (a better measure of actual time over target area than endurance which also includes flight time to target area) brings up 16 hellfire, 8 Small diameter bombs, a 30mm gatling, and a 105mm howitzer when you are tired playing games. It flies twice as fast and so can help troops under fire who may not have time to wait for drones with the speed of a Cessna 172. All with a superior sensor fit. Not Cheap at over 100 mill but much more useful.
The drones that are getting the serious attention by anyone not planning on retiring to a thinktank funded by Lockheed or General Atomics, are cheap disposable ones like the Switchblade which costs 100 grand but is a "suicide drone" that is crashed into it's target.
At least the Army is.
The Air Force is still dominated by pilots who want a "loyal wingman" --a drone controlled by a pilot--which will cost half as much a manned fighter but also have less payload and be one more thing for pilots to have to deal with.
"Loyal Wingman"
DeleteAs far as I'm aware, there is not a single exercise demonstrating the combat effectiveness of unmanned 'wingman' drones. The entire concept is codified wishful thinking and fantasy.
The report you cite is from an aerospace think tank whose avowed purpose is to push aerospace solutions and the report is written by two former AF officers. Shockingly, it describes an aerospace solution! On a seemingly unrelated note, are you aware of all the Navy spokesmen and officers who lauded the near-miraculous nature of the LCS? I didn't believe them, either.
The AF has acknowledged that large, non-stealthy UAVs are non-survivable on today's battlefield. If/when a stealthy, AFFORDABLE, COMBAT-EFFECTIVE UAV becomes available, I'll re-evaluate.
Feel free to cite a single, realistic exercise that demonstrates the combat-effectiveness of unmanned wingmen. I'll wait while you do.
Loyal Wingman is still in the concept development phase, so obviously no exercises have been performed. There probably have been simulations and wargame-level exercises. It does show the direction the AF is taking with unmanned aircraft.
DeleteRQ-170 and RQ-180 exist.
In 2021, the AF revealed the existence of two other classified UAV programs.
https://www.politico.com/news/2021/12/06/combat-drones-air-force-kendall-523812
I don't recall hearing anything else about them.
So their plans to move away from non-stealth UAVs are tied to programs like these. They aren't just moving away from non-stealthy UAVs to nothing.
At least I hope not.
Army aviation is one area where I could see unmanned actually providing some benefit.
DeleteCavalry operations are typically Recon and Security.
Recon is, of course, looking for the other guys.
Security is basically keeping the other guys from reconning you.
The recon role is someplace where I think unmanned can positively contribute.
A divisional cavalry squadron is limited in size and needs to be assigned to high value tasks.
Unmanned can help fill in the substantial gaps that can't be covered by the limited quantity of cavalry forces.
Unmanned also has the benefit of not risking pilots. A downed manned aircraft with pilots that need to be rescued is like sticky flypaper for a unit...unmanned not so much.
But there are limitations as well.
The unmanned world as seen by the operator is a limited-view two-dimensional image. The simple act by a pilot to turn his head to look around is more problematic for the unmanned. And the overall lack of situational awareness is a real thing.
The unmanned is also going to be less effective in engaging targets with weapons systems.
It's interesting that the Ft. Rucker commander compared the unmanned to Apaches, and unfavorably no less.
The Apache is a terrible scout platform. It's big, expensive, overpowered, overarmed, and with a large visual/audio/heat signature. I believe that the army became accustomed to using these for scouting in the low threat environments of Afghanistan and Iraq.
What makes more sense is to use pairs of armed MD530 aircraft for cavalry operations, with Apaches standing by to deal with threats that the scout birds can't handle.
The MD530s would be a fraction of the operating and procurement cost of the Apaches, which would tip the scales away from unmanned even more.
The best option, IMO, would be to use the unmanned platforms for general reconnaissance and situational awareness for the commander. This would be supplemented by the cavalry scout aircraft for high value, high-possibility-of-enemy-interaction missions, backed up by Apaches.
I would think that unmanned could work similarly for the navy, as a reconnaissance and situational awareness tool.
But I don't see many applications for unmanned taking over the warfighting roles from manned platforms without a substantial leap forward in technology.
Lutefisk
"Loyal Wingman is still in the concept development phase, so obviously no exercises have been performed."
DeleteAnd yet, reports indicate that the AF has already committed to a thousand drones (with no price point yet established, either!!!!!! that should work out just fine). So, not a single exercise with a prototype drone to demonstrate its combat effectiveness, if any (there isn't), and we're already into high volume procurement. Well, I'm not too worried. We committed to 55 LCS before we even had a finalized design and that worked out fine. We committed to 32 Zumwalts before we had a weapon for it and that worked out fine. We committed to several Fords before we had a working catapult, arresting gear, or elevators and that's worked out just fine.
I can't foresee any problems.
"They aren't just moving away from non-stealthy UAVs to nothing."
DeleteYes, they are moving towards nothing. They're moving away from existing, proven concepts to non-existent, totally unproven fantasy assets. That's nothing.
One may believe that drones can be made to work in combat but until there is even one iota of proof, it's just wishful thinking and not even solidly grounded, wishful thinking!
I'm all for exploring all alternatives in combat - ON A RESEARCH BASIS. Only an idiot would commit to a production run of a thousand units with zero proof of effectiveness.
I really want to hear how the lone pilot in a F22, F35 or NGAD is going to control 3 to 4 other aircraft on his own....I could see MAYBE a F15EX 2 seater, SUPER HORNET 2 seater or a E7 command control plane with multiple seaters control a couple of UAVs but some guy or gal by himself?!? And that's before the Chinese start jamming....once that happens, now what?
DeleteOnly response so far was that general saying the UCAVs are kind of like smarter "smart bombs" or better "cruise-missiles" than why do we need UCAVs then? Wouldn't be easy and cheaper just to make those a little bit smarter and not bother with UCAV? Why do we need the UCAV to be so smart, just make the bomb a little smarter and then UCAV then becomes a more simple bomb carrier if it's even needed?!?
The fundamentals that drive this paradox are:
ReplyDelete1. Software has high fixed costs. It takes a lot of people to write and maintain software. Manpower increases at low unit volume.
2. Software is brittle. New situations or high variance break it and requires human intervention. That intervention can be more than just having a human do the job in the first place.
For commercial software these problems aren't overwhelming because you sell a million copies and focus on very narrow cases. But there are still many examples like chemical plants where staffing went up even as automation software was added. Automation is known to increase labor in general assembly of car factories and companies like Toyota were the leaders in ripping out robots. ATMs didn't kill bank tellers, etc.
My blog's most popular post goes into detail on these topics and why we see so little economy wide productivity gain from software. I've also written that in the military context we should minimize what we expect software to do. There are only so many use cases with high volume and low variance that justify traditional software.
Machine learning is improving very fast right now. There is some possibility that it lowers the threshold for automation. But it can also make bonehead errors. These programs can beat the best players in the world at a complex game like Go, then an amateur can trick it using a strategy no human would fall for. As you say it is something for the R+D lab for now. It could very well change the world in less adversarial contexts while only being narrowly useful in military applications (drone navigation, etc.).
To be clear, there's a difference between program support manning (logistics, software, AI, facilities, etc.) and operational manning. This post and quotes are concerned with operational manning.
DeleteDefinitely, but it’s the same concept. A pilot in the aircraft is using all his senses to create a relatively basic output to move the stick or pedals, pull a trigger, etc.
DeleteFor a UAV you have to build software to relay all that information and process it from 1s and 0s to human friendly output. It often takes an unaffordable amount of work to create good enough software to replicate the information a pilot in the cockpit gets from his eyes or inner ear. So you have multiple operators managing sub par interfaces to make up for the brittleness.
So it’s even worse than it seems because you are paying more software and electronics engineers to create a digital analog that isn’t as good and requires more operational manpower. For this same reason I bet the F-35 would be a better plane relying on the pilot more and deleting most of its code, especially related to sensor fusion.
A civilian example is Tesla’s original general assembly line for its Model 3. They spent a huge amount of engineering resources and capital designing robots. But general assembly has a lot of tasks like installing wiring harnesses that are hard for robots. It performed terribly and they created a completely manual line in a tent in a weekend to supplement it. The tent line had less labor hours and defects from the start with basically zero CAPEX and planning.
These false economies of automation are part of how you get an LCS that has more labor than a Perry class frigate and costs more. Or you crash ships because you can’t afford software that is more intuitive than a physical throttle. The Navy would do much better relying on its sailors more using very simple interfaces. That would also remove tons of work in ship design and building, reduce maintenance, eliminate system interdependencies that worsen combat damage, and ultimately increase combat power.
Have read about the Navy's experimental ORCA program which has a goal of using an unmanned mini sub for mine laying and other applications. One wonders how this submersible would be recovered and about the cost !
ReplyDeletehttps://www.thedrive.com/the-war-zone/navys-85-foot-orca-unmanned-submarine-will-be-a-minelayer-first
The Dirty Little Secret of UUV's is they require a manned mothership submarine to bring from port to the combat theater. Thus, you will have to tie up one of the Nuke subs you are supposed to be freeing from mine duty.
DeleteIt is less dangerous than the manned sub laying mines itself, but still leaves the manned sub vulnerable as it waits for the UUV to return.
In theory, UUV's could be disposable depending on how expensive "disposable" is. The prootype Orca is about 10-15 million each if Boeing actually stays within the budget.
"ORCA"
DeleteAs of Apr 2022, the Navy was planning to terminate the program for a variety of performance, cost, and schedule failures.
I think we are blending LUUV and XLUUV here. The sub one, large, died. XLUUV is in the water. Its ride is a C-17, not a sub.
DeleteOrca IS the XLUUV and appears to be cancelled. The Large UUV (LUUV) is a glorified torpedo and the program is still functional, I believe.
DeleteNope
Deletehttps://www.nationaldefensemagazine.org/articles/2023/1/30/just-in-navys-first-extra-large-unmanned-sub-to-go-underwater-very-soon
https://seapowermagazine.org/navy-proposes-elimination-of-snakehead-lduuv-program/
You make a good point when you stated ; "UUV's could be disposable depending on how expensive it would be if lost". If the enemy recovers this submersible , then the technology would be exposed .
ReplyDeleteAnyone else reminded of Robert S. McNamara's efforts to push high-tech solutions THIRTY YEARS before the necessary technology is sufficiently mature to make these solutions viable?
ReplyDelete"Loyal wingman" drones MAY be able to replace human wingmen, and unmanned ships manned frigates as escorts; but IF they'll be able to (and it's still a very big IF), it'll be DECADES before these unmanned systems become capable of actually doing so.
So, why China is also working on unmanned systems, even on much faster paces than US.
ReplyDeleteAre they idiots?
"So, why China is also working on unmanned systems, even on much faster paces than US."
DeleteWho cares.
We need to work on our own problems, there's plenty there to keep us busy.
We need to have the best trained navy in the world.
Our ships need to be maintained in top notch condition and ready to go at any time.
We need to have ships and equipment that are properly designed for the specific jobs that they are expected to do in wartime.
We need to prepare for the right-now fight, not over allocate resources into the two-generations-from-now fight.
Lutefisk
Well said, Lutefisk.
Delete"Are they idiots?"
ReplyDeleteThe Chinese typically invest in things as cheap prototypes until they see if they work or not and then adjust their spending accordingly. They still publicly ANNOUNCE all of the things that they are looking at, however, as they and the rest of the world discovered long ago that they can "bleed" US R&D by just announcing that they are working on something and then watch the MICC salesmen and their uniformed employees spend vast amounts of money on "countering" them. Does it work? Oh ya - a single example - we are currently building a class of non-working aircraft carriers on the prayer that the Chinese will make the technology work since we can't.
In an attempt to understand the reasons why "unmanned systems require more manning than manned systems" (from your post above) I read the cited article (ref 1). I also read "The Dark Side of Unmanned Systems Autonomy" on defensemedianetwork.com. Beyond that I didn't find much on the internet to enlighten me. However, the statement in ref 1 that "[uavs] require even more humans to maintain them and to analyze the hours of sensor data they collect." I think could be key, regarding sensor data analysis. I don't know the percentage but I would guess that the majority of uav flights are used for gathering many hours of surveillance data - mainly video and radar - that must be analyzed by human operators for battlefield intelligence. This is currently a labor-intensive task. Is it not true that manned flights generally don't (and can't) have the same mission? I expect that video data analysis and video-radar data correlation is being increasingly automated as we speak.
ReplyDeleteDon't forget the UAVs' increased MAINTENANCE REQUIREMENTS. A UAV needs secure communications systems to relay commands from its ground control station, and to relay data to these control stations- commands and data a manned aircraft doesn't need- those communications systems need specialized maintenance from specially trained personnel.
DeleteDepends on what you mean by 'unmanned'-
ReplyDeleteIF you're looking for capabilities detailed by FX Holden in the Future War (MilSciFi) series, then- Yeah. We're a bit away from HIS loyal-wingman ideal.
OTOH.
CAPTOR is unmanned; doesn't require 24X7 supervision. MALD is unmanned; launch and leave (short duration). I submit somewhere between mines and Holden is what we should point towards.
Give me a missile barge with 200+ tubes, satellite controlled/directed and with a 10-inch freeboard (stealth). 200+ tubes controlled by one operator on a spot basis (launch order + target coordinates) is a force multiplier with no risk of lives.
Give me an unmanned sub with 8 tubes I can send in to the South China Sea. Exchange ratios being what they are.
USA has a history of exchanging money for blood and I agree with that calculus because I value lives more than money. See Russia as your alternative viewpoint/strategy.
I submit that rule-based systems (any target is a target; submersible targets get priority) are a more than reasonable start. Rule-based systems don't require constant manning. They can be timed-out or required to RTB before going Terminator on friendly forces.
And now that I've re-read this please allow me to modify.
DeleteI don't need 10 inches of freeboard. I can run my missile barge 10 feet underwater. If not more. Hell, surface the damned-thing and let it get destroyed. Post-launch. Lot of good that will do 'em.
Betcha' I could deploy a RoRo with no superstructure running 10 feet subsurface with a 1,000 plus tubes at a fraction of the cost of whatever the post-Ohio boat will be. Warheads on foreheads, quantity being a quality... And all that. Firepower. Firepower without exposing sailors. That's where I'm going.
Unmanned is the future.
Yeah, yeah we'll always need people but-
"Depends on what you mean by 'unmanned'-"
DeleteThere's not much debate about the meaning of unmanned in the current context. Unmanned refers to entire ships or aircraft, not missiles, mines, or decoys. Unmanned refers to equipment that is doing the job that a man used to do (sailing a ship, piloting an aircraft, driving a vehicle, etc.).
Re: unmanned. It really depends on the mission. The 135 man Grey Eagle company that the gene4al cirez can provide 24/7 coverage. It would take a lot more sorties, aircraft and personnel to do that with manned vehicles.
ReplyDeleteNote: this is not proponent of a UAV, but simply stating the comparison should he apples-to-apples.