Will We Have an Unmanned Armada?

Nov. 1, 2005

Unmanned aircraft will perform a remarkable range of missions over the next two decades. No longer mere reconnaissance platforms, these systems will be, in some cases, near-sentient robots able to carry out airlift, long-range strike, and even air-to-air combat missions.

That, at least, is the forecast and plan that is outlined in the Pentagon’s new “Unmanned Aircraft Systems Roadmap: 2005-2030,” released in August.

Robotic aircraft—formerly called “unmanned aerial vehicles” but now called “unmanned aircraft systems,” or UASes—are sure to elbow aside many manned airplanes that now conduct endurance missions of many hours; this has long been expected. However, the report’s confident prediction of more dynamic and autonomous roles suggests that a broad reshaping of the US military aircraft fleet may be in store.

In fact, the Defense Department may have been cutting back on manned aircraft programs—particularly fighters such as USAF’s F/A-22 and the joint service F-35—because defense leaders believe that equally powerful but cheaper unmanned replacements could be around the corner.

The roadmap explicitly forecasts the availability of unmanned fighter aircraft comparable to the F/A-22 and F-35 less than halfway through the expected service life of these new systems.

The document notes that UASes have been embraced by combat commanders at every level. Demand far surpasses the capabilities that current programs can produce. The systems have brought dramatic gains in knowledge of the battlefield and have “helped reduce the complexity and time lag in the sensor-to-shooter chain for acting on ‘actionable intelligence,’?” says the new study. The aircraft also offer new options for both kinetic and passive tasks.

Mushrooming Programs

At the same time, heavy demand has spawned a rush of development in all of the services. Mushrooming programs could easily end up producing redundant capabilities and wasting resources.

In the foreword, senior officials report that the roadmap is aimed at guiding the Defense Department toward a systematic migration of UAS mission capabilities, focused on the most urgent warfighter needs.

Those overseers included Stephen A. Cambone, the undersecretary of defense for intelligence; Marine Corps Gen. Peter Pace, then vice chairman (now Chairman) of the Joint Chiefs of Staff and head of the Joint Requirements Oversight Council; Kenneth J. Krieg, undersecretary of defense for acquisition, technology, and logistics; and Linton Wells II, acting assistant secretary of defense for networks and information integration.

The roadmap is addressed to the service Secretaries and heads of the Defense Advanced Research Projects Agency and National Geospatial-Intelligence Agency.

The 200-plus-page document includes a catalog of the myriad UASes now in the inventory as well as an appraisal of foreign systems either in service or in development. It also assesses the enabling technologies that are now emerging to allow UASes to be more capable and flexible. Such technologies include computer processing power and miniaturization.

The report also acknowledges some high barriers to achievement of the vision, barriers such as the reluctance of government officials to cede weapons release authority to a machine or, in the near term, to allow UASes to operate in civilian airspace. Another concern focuses on the reliability of the systems, which have had a tendency to crash.

As Air Force leaders see it, Pentagon officials should be cautious about rushing to push UASes into every feasible mission area. Lt. Gen. Donald J. Hoffman, military deputy to the assistant secretary of the Air Force for acquisition, said at a Capitol Hill UAS seminar in September that USAF doesn’t necessarily see a huge cost advantage to UASes in every role.

“If it’s big enough to hold a man, our analysis says to go ahead and put a man in it,” Hoffman said. That point of view is not driven by white-scarf mentality, he said, but by the cost trade-offs when a UAS reaches the size of a comparable manned vehicle.

Hoffman acknowledged that UASes can be made smaller, lighter, and cheaper by deleting the displays, ejection seat, and other man-specific hardware on an aircraft.

However, “what you save by taking out the physiological aspects, you have to replace with connectivity mechanisms” to ensure positive control over the vehicle and its weapons. For near-term applications, “it’s a wash; … there’s no cost driver” to eliminating a human, onboard pilot.

From UAV to UAS

The roadmap discards the term UAV in favor of UAS because such platforms must be connected to other elements for control and management as well as the dissemination of their intelligence “take.” Now in its third iteration, the roadmap for the first time discusses “near-space” aircraft.

In an August briefing for reporters, the deputy director of the Pentagon’s UAS planning task force, Dyke Weatherington, noted that the roadmap is “not a budget document” but a “technology roadmap.” It does not direct “anybody to do anything,” he said.

Nevertheless, he said the roadmap was built with the full participation of each service, and “there’s pretty good agreement on [its] goals and initiatives.”

The authors do not believe that a single entity should control or direct the Pentagon’s multitude of UAS programs. Weatherington had previously told reporters that selection of an executive agent for DOD’s unmanned systems would be premature.

Instead, DOD decided it would create two joint organizations—one for doctrine and concepts and another for the development of hardware. (See “Washington Watch: Sorting Out the UAV Situation,” September, p. 14.) The Air Force had sought to gain the status of executive agent for unmanned aircraft.

Now commanding some $2 billion a year of the DOD budget, UASes will account for about $13 billion in production funding through the end of the Pentagon’s six-year plan ending in 2011 and more than $1 billion in operations funding through the same period. By contrast, the Defense Department spent only $3 billion in total on UASes in the whole of the 1990s.

The roadmap puts UASes into four categories: Major, Special Operations, Small, and Unmanned Airships. Major UASes include the Global Hawk and Joint Unmanned Combat Air System. Special Operations UASes are unique to US Special Operations Command. Small systems are those that can be operated by one or two persons. The airships are aerostats or blimps.

In looking at the missions that might be performed by UASes in the future, the authors of the roadmap set out their “themes,” which clearly indicate the potential and reasoning behind what could be a broad recasting of the aircraft in a military context.

First, they say, UASes have matured to the point where they are no longer to be considered for “niche missions” but can be applied across the range of military activities. “Instead of asking, ‘Can we find a mission for this [UAS]?’ one will ask, ‘Why are we still doing this mission with a human?’?” the authors say.

Second, the authors seem ready to rely on commercial initiatives to meet future requirements; to accept capability that is delivered in installments, rather than all at once; and to think of UASes as disposable—which, so far, has not been the case.

Varied Solutions

“A 50 percent solution tomorrow is often better than a 70 to 80 percent solution in three years and better than a 95 percent solution in 10 years,” say the study’s authors. “Commercial solutions avoid using defense development dollars, which provides the opportunity for other developments,” such as thinking of UASes as short-term, “consumable” systems.

Rather than buying UASes in large lots, they might be bought in barely sufficient quantities and replaced or augmented a few years later with new and improved models. The authors make the analogy to TVs, DVD players, and desktop computers—items often cheaper to replace than to repair.

The authors want a thorough understanding of the mission to be performed before developing UASes to meet it, since they believe that UASes can be tailor-made with precisely the amount of capability needed—and not more.

“Do NOT,” the authors warn in capital letters, “make a [UAS] and then find a mission for it … [or] design a low observable aircraft and then try to figure out how to make it do a strike or suppression of enemy air defenses (SEAD) mission.”

Another emphasized theme is the view that progress in miniaturization will allow more and more capability to reside in a smaller and smaller package. The capabilities inherent in the original version of the Predator UAS, it is noted, now can be accomplished on the much smaller RQ-7 Shadow. Because of the high turnover in UAS development, it will not be necessary to build them with much growth capacity.

The final theme is that UASes “have the potential to solve a wide variety of difficult problems that may be unaffordable by trying to find solutions with traditionally larger platforms.” In other words, a UAS solution will be preferred automatically over the creation of a large new system.

In the near term, the Defense Department should focus on UASes in the roles of SEAD, strike, electronic attack, and intelligence-surveillance-reconnaissance, concludes the report. There should be a concerted effort to develop secure common data links to operate the aircraft and standardize their ISR products to be compatible with all weapon systems. Any system should be able to use data collected from any other UAS.

There should be a heavy push to develop, in the near term, safe and secure means of flying unmanned aircraft in civilian airspace, such that the Federal Aviation Administration is comfortable with their use. The unmanned aircraft should be given the ability to “see and avoid” other aircraft as soon as possible and also acquire the means to operate in adverse weather.

The roadmap notes that, in urban operations, geography can change quickly as buildings, scaffolding, and other obstacles appear and disappear in the course of days. A UAS onboard database would have to be refreshed constantly by a global geo-mapping function or risk being “useless” in an urban setting.

Commanders Want More

According to the roadmap, UASes are seen as a major response to meeting emerging needs of regional combatant commanders. When the commanders listed their top 50 “capability gaps” for the Fiscal 2006 budget process, “27 (54 percent) [were] capabilities that are currently, or could potentially be, addressed by UAS,” and the commanders specified UASes as the “desired solution” in at least four of those areas.

When polled, for all classes of UASes, combatant commander staffs rated “reconnaissance” as the No.1 mission for which they wanted unmanned aircraft, with precision target designation as the second or third most-wanted product.

The roadmap forecasts UASes supplementing or perhaps even replacing manned aircraft on the following mission timetable:

  • 2005-10: Communication relay and SEAD (replacing the EA-6B jammer with the Joint Unmanned Combat Air System, or J-UCAS).
  • 2010-15: Signals intelligence collection (Global Hawk), maritime patrol (the Navy’s Broad Area Maritime Surveillance system), and penetrating strike (with J-UCAS).
  • 2015-20: Aerial refueling and integrated SEAD/strike.
  • 2020-25: Surveillance and battle management (replacing E-3 AWACS and E-8 Joint STARS) and counterair missions (replacing the Air Force F-15 and F-16 and Navy fighters).
  • 2025-30: Airlift (replacing C-5, C-17, C-130) and integrated strike/SEAD/counterair (replacing Air Force F/A-22 and Navy F/A-18E/F).

In the last category, it is noteworthy that the currently planned 180 F/A-22s in the Air Force inventory will average fewer than 20 years in service when unmanned systems will (according to the report) be available to replace them. That is less than half their planned service life.

How will it be possible to replace fighter pilots with robots? The roadmap maintains that the human brain is “rated” at about 100 million MIPS (million instructions per second) and 100 million megabytes in memory and that processors capable of achieving this level of function are close at hand. The authors note that Moore’s Law, which anticipates a doubling of computing power every 18 months, means that computers will attain the processing level of the human brain in 2015, but “others estimate the memory capacity of a PC will equal that of a human memory closer to 2030.”

Moreover, “by 2030, the cost of a 100 million MIPS processor should approach $10,000,” meaning that an artificial, human-equivalent brain could become an eminently affordable “component” of a UAS.

The study adds: “As for inculcating a fighter pilot’s training and experience into a robot brain, the equivalent of Top Gun school for tomorrow’s J-UCAS will consist of a postflight download in seconds.”

Those processors will be vastly smaller than what we have today, the roadmap continues, and will likely be enhanced by “optical, biochemical, quantum interference switching, … and molecular (‘moletronics’) processors, or some combination of them, to achieve ever faster speeds and larger memories.”

The roadmap contains a series of assessments of state-of-the-art technologies, along with recommendations as to what areas should receive special funding and attention.

Bandwidth Bottleneck

In contemplating use of large fleets of UASes, officials worry greatly about the amount of electronic bandwidth available to send control signals to the vehicles and receive and distribute their sensor findings. The roadmap anticipates a self-healing solution for this problem.

“Eventually, onboard processing power will outstrip data link capabilities and allow [UASes] to relay the results of their data to the ground for decision-making,” says the report. “At that point, the requirement for data link rates in certain applications, particularly imagery collection, should drop significantly.”

Twenty years hence, UASes also may bear little resemblance to those of today, many of which would not look out of place at a radio-controlled-airplane enthusiast show. To provide lighter and stronger materials, the roadmap forecasts use of “transgenetic polymers” with “twice the tensile strength of steel yet … 25 percent lighter than carbon composites.” These aircraft also will be able to alter their shape in flight, constantly morphing to obtain the best speed, longest duration, or smallest radar cross section.

They also will be manufactured with microcapsules of “glue” that will allow the aircraft to repair itself if damaged in flight. This would not be a patch but a regeneration “to original condition.” The authors recommend investment in such materials research.

Likewise, the roadmap envisions antennas that can be “sprayed on,” eliminating the weight and power draw of existing systems. Human factors and simulation research also will make it possible for a remote human operator to receive an orchestrated variety of cues that will make it possible to have virtual presence in the UAS. “The future [UAS] pilot will transition from seeing the plane to being the plane.”

Power plants for future UASes will vary in power and complexity as much as the aircraft themselves. The roadmap suggests investment in a wide variety of propulsion mechanisms, ranging from scramjets and fuel cells to “reciprocating chemical muscles, beamed power, and even nuclear isotopes.” Photovoltaic (solar power) systems and reduced fuel consumption power plants, which can extend endurance and range 100 percent, are targeted for special attention.

Future UASes will be able to employ hyperspectral imagery techniques at long ranges, able to see beneath the surface of the earth and distinguish between types of vehicles and even people, following priority lists but able to exploit targets of opportunity as well.

Down to $1,800 Per Pound

A significant goal (and design driver) will be to cut the cost per payload pound that UASes can carry. Currently, it costs about $8,000 per pound for an average UAS to carry sensors. By comparison, the F-35 Joint Strike Fighter cost per payload pound is $7,300. On the J-UCAS, the goal is $5,500 per payload pound, and the long-term goal is to reduce the cost to $1,800 per payload pound across the UAS fleet.

Bobby W. Smart, USAF director for information dominance programs, noted that Global Hawk, as an example, costs “$40 million to $50 million a copy” just for the airframe, while the sensor package aboard costs $10 million. Because Global Hawks consume a great deal of manpower—needed in operations and maintenance of the aircraft and its systems, and people needed to collate its data, analyze, and disseminate it—the life cycle cost of some of the larger UASes “is not trivial,” Smart said.

Hoffman pointed out that UASes are made more costly and complex because of the need for “encrypted comms,” or unjammable communications with the aircraft. “You have to make sure only you can control it,” he observed, “especially when it’s armed.”

One of the most prominent challenges to widespread use of UASes will be frequency spectrum, Hoffman said. Within the US, “we’ve sold a lot of the frequency spectrum” to cell phones, satellite communications, high-definition TV, and the like. Broad use of UASes will require a new area of core competency, which Hoffman called “frequency dominance,” or getting other users off the frequencies when operating in a battle zone. While that’s possible over an enemy country, in peacetime exercises, “we can’t do it in [South] Korea,” for example.

Echoing the roadmap, Hoffman also noted that it has not been figured out yet how to safely deconflict what could be hundreds or even thousands of UASes buzzing around a battlefield. It will add cost “to add systems that will keep them from hitting things … or running into each other.”

In one of his last interviews as Air Force Chief of Staff, Gen. John P. Jumper in August said of the manned/unmanned question, “It’s hard to say what the right ratios are going to be.” He anticipates that long-range strike will be a natural UAS mission, as will ISR. He’s reluctant to give the air-to-air fighter mission, for example, to a UAS until it can be shown that it “doesn’t give up any of the quality we now have by having the greatest trained pilots in the world.”

However, if the technology bears out, he sees no obstacle to UASes taking on the fighter job, too.

“That’s the great thing about our system: Once it’s proven that it can do the work and it competes, then we’ll make those transitions if it’s appropriate.”