How can the USAF leaders write these things without risking laughter by their readers? It does not take a Billy Mitchell or Doolittle to see that the rise of UAV’s — unmentioned by General Deptula — begins a new cycle in air warfare, ending the dominance of manned fighters. Perhaps someone reading this either explain, or link to an explanation.
“U.S. Air Dominance Eroding“, DoD Buzz, 15 September 2009 — Excerpt:
The U.S. military’s historic dominance of the skies, unchallenged since around spring 1943, is increasingly at risk because of the proliferation of advanced technologies and a buildup of potential adversary arsenals, according to Air Force Lt. Gen. David Deptula, the service’s chief for intelligence, surveillance and reconnaissance. Speaking today at the Air Force’s annual convention in the Washington area today, he provided a wide ranging assessment of what the QDR team is calling “high-end, asymmetric threats.”
Emphasizing the increasing capabilities of “anti-access weapons,” such as long range precision missiles, Deptula said pilots in future wars will not operate in the “permissive” threat environments of current wars in Iraq and Afghanistan. Deptula, best known for crafting the Desert Storm air campaign, said potential opponents have learned from U.S. operations and will use precision arsenals to stop a buildup of U.S. airpower near their borders before a war even begins.
… Because of improvements in over the horizon and passive radars, U.S. aircraft will be detected long before they reach their targets. “The area that we operate in free from detection is rapidly shrinking,” Deptula said, “our adversaries are going to have capabilities that we’ve never operated against.” The newest generation surface-to-air missiles, such as the Russian SA-21, have ranges exceeding 300 miles and the ability to target low flying aircraft, and will likely be exported.
Speaking to the more traditional realm of air-to-air combat, so dear to his audience’s heart, Deptula contends that the U.S. technological edge there is eroding. While “fourth generation” fighters are no match for the most advanced U.S. fighters, Deptula reminded the audience of the Russian export success with the MIG-21, some 12,000 of which were built, and operated by over 50 countries.
Russia and China are both developing “fifth generation” fighters that will be widely exported at prices that will undercut the F-35 price tag. Both nations will thus acquire “near F-22 performance… while attempting to proliferate the [aircraft] to perhaps near F-35 like quantities,” he said. “We may be facing a fighter threat capability in quantities we’ve never experienced before.”
UAV’s have — or eventually will have — other advantages over manned fighters:
- Smaller signature, making detection more difficult.
- No human crew onboard to limits its ability to maneuver.
- Lower costs, allowing construction and operation of more aircraft.
- Lower cost and no crew allows taking more risks.
The last might be the most important. The increasing cost of manned aircraft and the ever-lower willingness to take casualties perhaps limits the use of manned aircraft more than any other factor. Warcraft are useless unless they can be put in harms way. UAV’s can be expended like ammo, assuming the targets are cost-effective.
UAV’s lower construction and operating costs, plus the far lower cost of training operators, once again opens the context for command of the sky. UAV’s will move along the characteristic declining cost- increasing capability curve of modern tech, allowing even a medium-sized nation to deny its sky to US manned aircraft.
A look into the future
“The Last Manned Fighter: Replacing Manned Fighters with UCAVS“, Robert B. Trsek (Major, USAF), Air Command and Staff College, April 2007 — Excerpt from conclusion:
The age of UAVs is upon us, and as technology advances exponentially, the Air Force must decide what the next fighter will look like. How will we replace the F-22 and F-35 twenty or thirty years down the road? There are no technological barriers that prohibit the design and use of UCAV fighters on a large scale. They are capable of cooperative employment, air refueling, WVR and BVR engagement, AI and CAS. There are anticipated limitations in bandwidth and concern for performance in WVR if MITL is the ultimate solution to command and control, but this can be overcome with automated sequences once permission to engage has been authorized.
The range and endurance of UCAV fighters offer persistence and attractive options in a world of growing anti-access strategies. They offer advantages in performance, altitude, and employment without the limitations of human physiology. UCAV fighters deny the political use of POWs by our adversaries and deny them a great tool in a media campaign.
Ultimately, UCAV fighters are not a panacea, but offer the presence of force in a threat environment that twenty years from now will be extremely lethal. The costs and risks associated with UCAV fighters are significant but surmountable. The single point of failure may be in our command and control through the RF spectrum for over-the-horizon operations. If sufficient bandwidth can be secured, and the control of remote vehicles can be assured, there are immense dollar costs to be saved in their employment. While we cannot place a price on the human element, the value of our pilots is demonstrated in the training they receive and the assets assigned to recover them.
In the context of future threat systems and anti-access strategies, the Air Force would be foolish not to pursue UCAV fighter technology.
Victory smiles upon those who anticipate the change in the character of war, not upon those who wait to adapt themselves after the changes occur.
— Giulio Douhet, ‘The Command of the Air.’
George Friedman of Stratfor said that “The Air Force, which was built around the concept of air superiority and strategic bombing, has a visceral objection to unmanned aircraft.” In the comments Duncan Kinder expanded on this and pointed to a solution:
There is a poetic / psychological dimension to this. The fighter pilot is a mythic figure dating back to the Bloody Red Baron. As such, he captures one’s imagination.
The trick is to find a poetic substitute. To me, the image of falconry works. The idea of sending birds of prey to swoop down upon one’s target jives with the UAV.
Some Articles about military UAVs
- “Unmanned Aerial Vehicles/Unmanned Combat Aerial Vehicles – Likely Missions and Challenges for the Policy-Relevant Future“, Manjeet Singh Pardesi, Air & Space Power Journal, Fall 2005 — A contrary (or conservative) perspective.
- “Unmanned Aircraft Systems Flight Plan 2009-2047“, UASF, 18 May 2009
- “US now trains more drone operators than pilots“, The Guardian, 23 August 2009
- “AF Intel Head Embraces 4GW“, Chet Richards (Colonel, USAF, retired), Defense and the National Interest, 20 September 2009
- Fighter UAV’s, an excerpt from The Spectrum of Future Warfare by Carlton Meyer (Captain, USMC, retired) — Vivid, although the author is not an expert in this field.
- Wikipedia entry about Unmanned Combat Air Vehicles and the History of Unmanned Combat Air Vehicles.
Afterword
Please share your comments by posting below. Per the FM site’s Comment Policy, please make them brief (250 word max), civil and relevant to this post. Or email me at fabmaximus at hotmail dot com (note the spam-protected spelling).
For more information from the FM site
To read other articles about these things, see the following:
- About America’s national defence strategy and machinery
- About Military and strategic theory
- About Naval warfare and strategy
Reference pages about other topics appear on the right side menu bar, including About the FM website page.
The F-16 was known as an aircraft that could outperform the pilot, i.e. the plane could perform maneuvers that would incapacitate or kill a pilot. Seems to me a swarm of highly maneuverable unmanned robotic aircraft could sweep the skies of human operated aircraft but hey I was just an enlisted puke. Also, how come the Russians and the Chinese can be a fleet of “fifth generation” fighters to tangle with Deptula’s F-22’s & F-35’s at cut rate prices? Maybe we should ditch the F-35 and buy their version instead, especially if you get almost the same performance for half the cost.
In the distant and asymmetrical future, our ability to interdict afghan wedding parties might be substantially reduced?
.
.
Fabius Maximus replies: Probably not. That is, great and mid-sized powers might be able to defend their airspace. Afghanistan is only a nation by the peculiar rules of our time, but not in any meaningful sense.
There are two enormous hurdles for the unmanned future:
– The C2 challenge is enormous – cyber warfare will completely change optimistic assessments of assured bandwidth and reliable communication links in the air.
– Artificial Intelligence necessary for intercept is still decades away. UCAV can replace future strike operations when C2 isn’t impeded, but the amount of software necessary for autonomous intercept is still not on the horizon. I do not believe the programming language necessary for autonomous intercept capable of replacing a human in that role has been invented yet, nor the core hardware architecture.
No remotely piloted vehicle in the world can intercept a EA-18G or similarly capable platform, which suggests an enemy strategy should we attempt remotely piloted intercept vehicles.
.
.
Fabius Maximus replies: While your time forecast might be correct (“decades away”), the history of the past 25 years suggests that this might happen far sooner than expected. I would like to see some expert analysis of this question.
The use of the term “artifical intelligence” is somewhat prejudical. In practice it means technology that has not been invented. When we have software with these specific capabilities, it will no longer be “AI.” Example: speech recognition. Once sci-fi, a key goal of AI. Now routine, but no longer considered AI.
There are additional cost savings here as well. Jet fighter pilot training is EXPENSIVE. On the other hand, UAV missions can be flown successfully by high-school dropout, Xbox fanatics.
In the futuristic movie “Rollerball” there’s a scene where drunken elites play around with a pearl handled six shooter. Though clearly not a military weapon, the scene reveals the handgun has roughly the destructive potential of a Howitzer, making the point that widespread war is definitely something to be avoided. We’re not there yet, but we get closer every day.
I honestly think it’s going to be more problematic than people think to get extremely advanced unmanned systems fielded, but even if they do become widely operational, so what? They don’t represent a new strategy for dealing with the world, just another step on the weapon/ counter-measure ladder. If anything, their deceptive ease-of-use could get us involved more often in unfavorable conflicts.
Given the past trends on new technologies, its highly unlikely that autonomous fighters will magically be delivered to the airforce problem free and on time. The overall complexity of the aircraft software and AI software is going to be problematic and one which isn’t likely to be solved easily. While it might be that in a decade or so we have an AI which can fly an aircraft to some degree, its going to be a while longer before its capable of flying combat missions with the flexibility of a human pilot.
The USAF has plenty of other acquisition problems to overcome before it can worry about autonomous aircraft. The tanker replacement program has become a fiasco, the F-35 project is well behind its testing schedule, will likely be overbudget and will be outclassed by future Russian SAMs and possibly even future non-stealth fighters.
.
.
Fabius Maximus replies: The use of the term “AI” just clouds the discussion. Current autopilot technology can handle take-off, flight, and landing — no “decade or so” is required. Also, the development of UAV fighters might not be led by the US. It’s a technology potentially capable of ending the USAF’s control over much of the world’s sky, and so within the capability of many other nations to develop it.
Just as US intelligence consistently underestimated the time required for Russia, China, and India/Pakistan to build an atomic bomb, so I suspect we underestimate the time required for UAV’s to challenge the USAF for control of the sky.
The essential point Deptula makes is that despite our investment in offensive airpower, air space control can be effectively exercised by defensive forces using networked systems. The historical data supports this assertion: Dowding’s radar controlled fighters during the battle of Britain, Kamhuber’s Luftwaffe night fighters during WW2, N. Vietnam’s SA-2/Mig/Ground Control, the devastating lethality of Arab netted SAM/AAA during Yom Kippor (which according to Israeli sources won air superiority until overwhelmed by Israeli ground forces), and perhaps more recently, the rumored scattered successes of the Georgian air defenses against Russian air assault last year.
When networked defenses are operating with roughly equivalent technology to attacking air power, the defense can generally raise the loss rate beyond the attacker’s sustainable limit. The main thrust of Deptula’s angst is that we are facing this precise situation…and it will rapidly become a strategic limitation on US military employment when US airpower cannot assure its usual level of success.
The UAV discussion is a sidebar: UAV’s offer, above all else, persistence. The most useful employment of UAVs is in armed reconnaissance and suppression of enemy air defenses, both of which require loiter. However, both missions require real-time control to defeat increasingly rapid enemy engagement cycles, and this is problematic (not impossible) for the current C2 level of capabilityvwhen UAVs operate in contested airspace.
I caution everyone on UAV capability: it is quite basic in many ways and is not prepared to take on sophisticated missions, nor will it meet expectations for onboard data processing and decision making for another technological generation…and it will be very expensive.
Although no longer an expert, I have experience with the F-4G, EA-6B, F-15E, F-16, and EA-18G platforms (cockpit for all except the Navy EA-18G, for which I have high hopes), as well as UAV developmental experience with the Global Hawk, Dark Star, and Boeing UCAV demonstrator (primarily flight test and operator). My experience was unique and will not be easily found in the USAF today which bet (years ago) that investment in stealth and black magic could replace the systems and airmen who focused on defeating enemy networked systems.
Deptula’s commentary is worrisome, especially given his relevant history as an air campaign planner and his current intelligence portfolio. Reading the tea leaves with the knowledge of how we got here, the investment bet appears to have been lost. We may have an expensive yet diminished airpower capability relative to our very, very high expectations.
>Russia and China are both developing “fifth generation” fighters that will be widely exported at prices that will undercut the F-35 price tag. Both nations will thus acquire “near F-22 performance… while attempting to proliferate the [aircraft] to perhaps near F-35 like quantities,” he said. “We may be facing a fighter threat capability in quantities we’ve never experienced before.”
They can’t just say they need a pile of money to counter China and Russia – because the next question is how many cities will the US be prepared to lose in such a war. And they can’t just say you need the money to counter countries like Iran because they are too weak to be a realistic threat. So you get the mixing of the two ideas, the size and money of China with the possibility of a war with Iran.
When they squint really hard they can about see the Cold War again.
The ridiculous military oscillation between bravado about how great they are, and running around like scared little girls telling everyone that they are doomed unless more money is poured down the hole should just be laughed at.
The reality is that large cuts should be made in the US military. It’s a bloated dinosaur with an army that cant defeat a couple thousand farmers, a navy that cant defeat a few tens of pirates and an airforce hasn’t had anybody to fight for 30 years an is hopelessly lost in the past.
But it can’t happen because the American public believes that more power is always better. So they will continue to be played for suckers by the defense establishment.
A point not yet brought up – UAV’s dependence on satellite architecture. China needs no “near-F22” to blunt U.S. air power. They already have missiles capable of destroying U.S. military satellites; what else is needed?
.
.
Fabius Maximus replies: Cannot UAV’s be operated locally, without relying on satellites? And precise navigation was possible before GPS, using other electronic navigation aids and inertial location tracking.
I believe T. Greer refers to this section of the speech:
The discussion regarding UAV vs manned aircraft in dogfights misses the point. Dogfighting is the least reliable, costliest, and riskiest method of establishing air superiority or air dominance. It is akin to establishing hornet dominance by swatting individual hornets. Better to destroy the nest by establishing conditions which leave the enemy air force on the ground.
The USAF’s obvious achilles heel is the huge, expensive, vulnerable, and hard-to-replace network of fixed bases necessary to put their aircraft in the sky. When Deptula refers to anti-access technologies, he is talking about attacks on USAF bases. If the enemy destroys a USAF airfield with a swarm of cruise missiles, for example, it doesn’t matter how superior the F22 or F23 are or how many we have.
This is a long-standing trend in warfare-as as weapons grow more accurate and deadly, the forces facing those weapons must disperse. In the face of widespread precision guided munitions, it is arguably better to invest in large numbers of cheap, decent weapons rather than in small numbers of wonder weapons. However, the US acquisition process favors wonder weapons, as they are more efficient at their secondary (primary?) purpose of funneling money into key Congressional districts.
.
.
FM Note: Andy refers to this section of the speech:
@Guy – post 8: If you were in charge, how would you configure our airforce?
I suppose there are two ways you can view UAV’s: As a new improvement in warfighting like the introduktion of tanks or submarines into warfare a hundred years ago. The potential for improvement is almost unlimited just like we continue to develop tanks and submarines into ever increasing variants. Or as a dead-end – a promising avenue that ends in failure. Warfare is full of “silver bullets” that turned out to be disappointing when faced with a real challenge. Do people remember all the hype about RMA ten years ago? Or for that matter the promise that strategic air warfare could win wars “from the sky” (like Douhet promised a hundred years ago)?
I feel it is necessary to mention the last thing because there is a strong possibility that UAV’s won’t live up to expectations when they are faced with an adversary who can actually fight on equal terms. There is a difference (although for some reason many don’t understand that) in bombing people with UAV’s in Pakistan or fighting – say – the Chinese army across the Taiwan Strait. We might be up for some nasty surprises and I have to mentioned that already in the sixties there were predictions of the “death of the fighter pilot” due to SAM’s. They proved premature. While I myself am confident that UAV’s have gained a place in modern warfare it still has to be determined exactly what place: As a supplement to manned aerial warfare or its replacement.
The truth is we don’t really know what the future of warfare will mean. I can only guess that our expectations and beliefs will have to be abandoned just like many expectations or beliefs about warfare had to be abandoned when Europe went to war in 1914. One thing is to shoot at natives in Africa with a Maxim gun with impunity, something quite different to get shot at yourself with a Maxim gun (not to mention shells and poison gas).
From a technological perspective, if we don’t have the hardware/software to enable autonomous targetting and fire control for UAVs within the next 10 years at the latest, I’ll be quite surprised. Remember, it doesn’t have to be some walking terminator with advanced reasoning capabilities. It has to:
1) Take off (can do already)
2) Fly to a preprogrammed point (can do already)
3) Acquire targets and fire
4) Return when certain conditions are reached (time, ammo, whatever)
5) Land (can do already)
Doing point 3 with “SCI-FI” levels of precision is very difficult. Doing point 3 with “turn on radar, shoot the blinkies” levels of precision is a trivial problem already solved by much of our smart weaponry.
The key is to not think of them as ‘robotic planes’. Think of them as ‘smart missiles’ with or without submunition capability, and you’ll have a better understanding of the problems posed. Remember, if the UAV is cheap enough and its targets are worthwhile, options 4 and 5 are optional :)
The above comment refers to UAVs in a ‘aerial denial’ role. Obviously, once you’ve cleared the skies, you’re free to use your more valuable aircraft for things that require more thinking/precision.
More on pessimism about technological progress
The repeated mentions of AI in this thread are misleading.
As Fred Reed wrote about people’s perception of software progress: “Doesn’t count as intelligent: It isn’t perfect, and anyway I’m used to it.”
AI researcher Rodney Brooks: “Every time we figure out a piece of it, it stops being magical; we say, Oh, that’s just a computation.” (source)
Tesler’s Theorem: “AI is whatever hasn’t been done yet.”
From Pamela McCorduck’s Machines Who Think (2004): {It’s an} “odd paradox {that} practical AI successes, computational programs that actually achieved intelligent behavior, were soon assimilated into whatever application domain they were found to be useful, and became silent partners alongside other problem-solving approaches, which left AI researchers to deal only with the ‘failures,’ the tough nuts that couldn’t yet be cracked.”
From Ray Kurzweil’s The Age of Intelligent Machines (1990): “AI research has made enormous progress in only a few decades, and because of that rapidity, the field has acquired a somewhat shady reputation! This paradox resulted from the fact that whenever an AI research project made a useful new discovery, that product usually quickly spun off to form a new scientific or commercial specialty with its own distinctive name. These changes in name led outsiders to ask, Why do we see so little progress in the central field of artificial intelligence?”
These references are all from Wikipedia’s entry on the AI Effect.
Reply to 12: A USAF that can not gain dominance over contested airspace will not contribute meaningfully to military decision. The current (and legacy) warfighting concept of “precision targeting without loss or attrition” is not technologically supportable when the enemy has equivalent technical means (such as currently fielded SAMs in netted defenses). Taken together, these two ideas lead one to the conclusion that the USAF must realize that is in danger of being ineffective due to the loss of its technical superiority AND the due to the investment in a presumed superior technical force that is too small in size to absorb loss.
If my assumptions are accepted, the USAF needs to invest in air dominance systems that it can afford to lose. This is probably a manned/unmanned force mix and employment concept that can effectively hold enemy high value defensive weapons (a modern long-range SAM is very costly) at risk.
Again accepting Fabius’ comments on the realization of technologies mimicking intelligence, the “needs to happen” technology is high confidence automated target recognition. One immediately is aware that the various airborne sensor types (radar, IR, EO, electronic signal) do NOT produce the level of certainty typically required to launch weapons within our current legal rules of engagement context. Even if these rules were liberalized, a machine has great difficulty in discriminating/identifying targets based on airborne sensor data.
My summary then is invest in attritable air dominance, create a doctrine that realistically accounts for combat loss, and develop target recognition technologies that support these first two initiatives. One should note that target recognition is a field with a thirty year history of incremental success despite heavy investment.
Fabius Maximus replies: The use of the term “AI” just clouds the discussion. Current autopilot technology can handle take-off, flight, and landing — no “decade or so” is required.
You are missing the point. AI is at the very center of any unmanned technology evolution that takes unmanned vehicles from remotely piloted robot systems used over uncontested battlefields to autonomous vehicles of war by a contesting peer opponent. The complexities involved in the AI for the evolution you are discussing are beyond our technology today. It is science fiction without technological breakthroughs including a more efficient programming language that is more adaptive (an area many Universities are focused on in terms of AI development).
How can you have a realistic discussion of unmanned vehicles replacing manned fighters if you simply accept the premise the single greatest technology hurdle is assumed overcome?
Accepting the premise that we will achieve realization of technologies mimicking intelligence forces the answer to the question FM raises based on General Deptula’s comments. If you accept the premise, then manned systems will not be needed in the future. If you don’t accept the premise, which most Navy and USAF people do not, then General Duptula is right.
I don’t know anyone in the defense industry or Pentagon who casually accepts that premise that AI can be assumed in the future, indeed I have never heard it outside of a science and technology conference. The AI challenge is enormous, because the reason both the Navy and USAF want to slow down on existing unmanned technologies (nevermind future tech) for high-end asymmetric threats is because neither service is fully confident they can overcome C2 challenges with current (and planned) C2 technologies to effectively operate remotely controlled unmanned vehicles in contested space. Both services have long pointed to several necessary capabilities that have not been fielded (not fully developed), like C2 oriented Operational Responsive Space (ORS) for example, just to get more out of existing technology in unmanned systems.
The Northrop Grumman UCAS is a perfect example. Northrop is planning enormous redundancy including technology integration with E2-Ds, JSTARS, and other battlefield systems just to build in some communications redundency because they have no confidence in the constellation systems ability to communicate with a UCAS in contested spaces. There are huge bucks saved on pilot training, but those costs are simply transferred to creating a redundant C2 network… and over contested space, I don’t see E-2Ds or JSTARS being sent too close to enemy defenses… so even that plan raises questions.
.
.
Fabius Maximus replies: You describe an evolutionary process as a binary outcome — overcome or not overcome. The history of AI shows that it is a process of incremental progress, which in this case will be gradual increase in the capability of UAV’s. As you well know, technological progress has consistently been over-estimated in the short-term, and underestimated in the long-term.
To state this differently: how many of the industry and Pentagon analysts in 1990 correctly forecast the extent of UAV capability we have in 2009? Not many, I expect. Their consensus forecast for 2030 probably will fall short by an equal margin, IMO.
Captain Kirk has a cell phone, considered 23rd century technology by the writers of 1968. While ours are not as powerful as his, I doubt it will take centuries for them to be developed.
Go back to square one, please. Who are these adversaries who will challenge our dominance of the sky? This whole discussion is soooo twentieth century. Deptula’s real question is how soon UAVs will extend their role to encompass all or most of those traditionally carried out by manned aircraft. And I agree the answer is, “soon.”
.
.
Fabius Maximus replies: Iran. North Korea. And the next Iraq.
@Guy – post 16 – thank you
I’d have thought Goliath ‘s UAVs were very vulnerable to David’s stones .
Ha ! Got the name. Live-Shot.com , and live-paintball .
Probably all from the creator of Bonsai Kittens.
Maybe Live-drone-USAF soon available on internet . Who has the fastest broadband to turn the thing round ? Install boomerang.co.alqueda for a small weekly subscription !
So many experts propounding “truths” about air power.
1. Fixed bases. Sure they’re vulnerable. But ask the Marines if they really operate their Harriers (and will operate their F-35 STOVLS) from open fields. Hint: Think about logistics. But if you use aerial refueling and the other guy doesn’t, your bases are out of range. As usual, the answer is, “It depends.”
2. The unlikeliness of state warfare, especially between nuclear powers: It’s pretty hard to see us going conventional with China, but it’s not fantasy. A crisis over Taiwan could trigger it. Also, nukes tend to have a restraining effect, but they don’t prevent proxy wars. Britain didn’t think they needed aircraft carriers until the Falklands. The trick is to plan for the most likely events 4GW and hedge the risk. Today the system does neither very well.
4. “UAVs will replace piloted aircraft in the air supremacy role.” Lots of technological hurdles, but UAVs may be a flash in the pan, too. Airborne laser is testing this year. Tough to maneuver even at 20g against a beam of light. The point is, my crystal ball ain’t much better than yours.
No one would claim they could command an Army division without the requisite expertise, but somehow, everyone is an expert when it comes to employment of air power. “It’s easy, all you have to do is x, y, and z.”
#22 , yes , choked on my cuppa tea when heard announcement as to Reason 17(a) why Britain couldnt send more helicopters to Afgh – ” Nowhere to park them ” . Actually made sense after some thought .