Can an innovative business model ensure the best chance of success for the US Department of Defence's ambitious plans for armed autonomous aircraft?
X-plane programmes are intended to explore the unknown, but the barriers to be broken are usually technological. The USA's Joint Unmanned Combat Air System (J-UCAS) programme, already vaulting technical and operational hurdles, now faces the daunting programmatic challenge of pioneering a new way of doing business.
Hardly surprising, therefore, that the J-UCAS programme is back under the control of the US Defence Advanced Research Projects Agency (DARPA), which prides itself in being willing to explore the "far side" and adopt unorthodox approaches in pursuit of its vision. "We recognise we do not have all the answers," says programme director Dr Michael Francis. "That's why DARPA is back in charge."
Unmanned combat air vehicle (UCAV) development may have begun at DARPA, but it was diverging in US Air Force and Navy directions when the Department of Defence ordered the efforts to be combined under the J-UCAS programme and placed back under the control of its research agency.
Early this month, DARPA plans to seek bidders for the role of integrator/broker on the J-UCAS. The job is unusual, but crucial to the agency's plan to create a consortium to develop the J-UCAS in a manner that maintains competition between the prime contractors, Boeing and Northrop Grumman, while encouraging other companies to come forward with technologies. "If this was just a hard engineering task, we are confident it would work, but we are pushing technology and not sure either one has all the best ideas," says Francis.
The result is a programme structure distinctly different to previous successes. The Northrop Grumman RQ-4 Global Hawk unmanned air vehicle was a single-service technology demonstrator that evolved without competition into an operational system. The Lockheed Martin F-35 Joint Strike Fighter was the winner of a joint-service procurement following an intensely competitive concept demonstration and technology maturation effort.
J-UCAS is not a procurement programme, but is more than a technology demonstration, and could have many different outcomes, says Francis. With DARPA back in charge, the services have taken a step back and will not begin to finalise their requirements until the operational assessment phase planned for 2007-9 begins.
When Boeing won the DARPA/USAF UCAV contest in March 1999, its X-45A technology demonstrator was intended to evolve into an operational system without further competition. This was underlined in August 2002 when Boeing received a $460 million extension to its original $131 million contract to build the larger X-45B, aiming for an initial operational capability with the US Air Force by 2008.
The US Navy, meanwhile, had elected to maintain competition for the naval UCAV-N, with DARPA awarding study contracts to both Boeing and Northrop Grumman in June 2000. But Northrop Grumman had to build and fly its X-47A UCAV-N demonstrator on company funds to stay in the race. That changed in April last year, when DARPA launched the Joint Unmanned Combat Air Vehicle (J-UCAV) demonstration programme.
Under the J-UCAV programme, Boeing was instructed to abandon design of the X-45B and build a minimum of three even bigger X-45C demonstrators, while Northrop was funded to build at least two X-47Bs, much larger than its X-47A. Renamed to reflect an increased emphasis on the overall system, rather than the individual vehicles, J-UCAS became a separate programme office within DARPA in October last year.
Now a $4 billion joint programme, J-UCAS was structured as a series of development spirals, each with greater capability. The A-model X-45s and X-47 became Spiral 0, tasked with demonstrating the technical feasibility of autonomous operations. The initial, Spiral 1, X-45Cs andX-47As were intended as risk-reduction vehicles, focused on demonstrating survivability and carrier suitability ahead of Spiral 2, the operational assessment.
As aircraft, the X-45 and X-47 still reflect their service-specific origins. Designed for USAF suppression of enemy air defences (SEAD) and strike missions, the X-45A is a small, survivable vehicle intended to be stored and transported into theatre in a container. The bigger X-45B grew out of the requirement to perform deeper strikes and electronic attacks against integrated air-defence systems, while the even larger X-45C recognises the lack of access to overseas bases and is designed to operate from the continental USA.
Different in planform to the X-45A and B, the C model incorporates features from Boeing's X-46 UCAV-N design. Its size is also close to that required to fulfil the US Navy's priority mission for J-UCAS - persistent intelligence, surveillance and reconnaissance. But the X-45C is still smaller than Northrop Grumman's X-47B (see table), which was designed from the outset to meet the USN's requirements, which include aerial refuelling and carrier capability.
"Northrop Grumman's focus was on UCAV-N; they had to go create a land-based SEAD vehicle. Now both contractors are working on both services' requirements," says Francis. But the two demonstrators are just the starting point for J-UCAS, he says: "The X-45C is not great for carrier demonstrations and the X-47B does not meet all the low-observability goals."
Earlier this year, the J-UCAS programme was restructured to create a third architecture element - the common operating system (COS) - to connect the two different air vehicles. "Anything beyond the platform is in the COS; anything that connects the platforms is part of the COS," says Francis. "We have elevated the COS to a separate system."
The result is a dramatic redirection of the programme, which DARPA hopes to complete by mid-July. Spiral 1 and 2 have been amalgamated. "There is not enough time. The luxury of Spiral 1 risk-reduction has gone," says Francis. Controversially, the two prime contractors are being asked to abandon their different operating systems and contribute to a common solution that allows not only their own UCAVs to "plug and play", but also other future vehicles.
Differences in the designs, and in the scope and timescales of the original UCAV and UCAV-N efforts, is one rationale for restructuring J-UCAS, as it levels the playing field for the two primes. "Boeing has been working UCAV software for some time, Northrop Grumman has a great track record in UAVs; putting them together is good," says Francis. The new consortium structure "also allows for other ideas, and gives us the best", he says.
Industry reaction to the restructuring has been mixed, and negotiations on forming the consortium have gone slower than planned. While privately expressing reservations about aspects of the scheme, both Boeing and Northrop Grumman publicly support the common operating system. "We are all fully supportive of the concept and the vision," says Mike Heinz, general manager of Boeing Phantom Works' Integrated Defence Advanced Systems. "The devil is in the details."
The concept of a common system with an open architecture is a "good idea", says Scott Winship, Northrop Grumman J-UCAS programme manager. "But if you take it to the extreme, with common everything, where is the innovation and competiton?"
Although DARPA's stated goal is to develop a single suite of sensors accompanied by common mission avionics and software applications, "we are asking the contractors what makes sense to be common", Francis says. "Commonality only makes sense if it works. If it's a force fit it will make them uncompetitive."
For DARPA, the restructuring of the J-UCAS programme is a return to the roots of its work on unmanned combat air vehicles, and to the Uninhabited Tactical Aircraft (UTA) project that ran from 1994 to 1996 and preceded the UCAV and UCAV-N programmes. From DARPA's perspective, the UTA's combination of unmanned aircraft, information technology and human decision making was an "erector set" for network-centric warfare.
Creation of the common operating system for J-UCAS is a return to the concept of UCAVs as a system of systems, rather than a collection of vehicles. The UTA initiative focused on using groups of UAVs collaborating to provide innovative ways of attacking ground targets in heavily defended areas. The air vehicle was to be small and low cost, and DARPA's focus was on using networks to enable the concept.
The air vehicles may have grown in capability and cost, but the network has re-asserted its importance in the latest iteration of the J-UCAS programme. The common operating system is the software that provides for the command and control, communications management, mission planning, interactive autonomy, human/systems interface and other functionality required to enable the UCAVs to operate with each other and with manned aircraft.
Francis uses the analogy of a desktop-computer operating system (OS). "The OS evolves at its own pace, and at some point you upgrade your hardware. The COS is all software and we can let it evolve at its own time constant, which is faster than the hardware will change."
Conversely, it will be possible to make air-vehicle changes without impacting the operating-system development cycle. "If we separate them we get much more flexibility, we can accommodate change more gracefully, and deal with risk issues in development without dragging the whole programme down," says Francis.
By creating the consortium, and making the primes work together and bring in other technology providers, DARPA hopes to get the best software application for each operating-system function, including:
resource management, which includes managing sensors and weapons; information management, which receives, publishes and fuses data; battle management, which creates a common operational picture and provides integrated fire control; contingency management, which enables the system to handle failures and complete the mission or return intact; network management, which allocates bandwidth and maintains service quality over datalinks ranging from secure satellite communications to line-of-sight networks; mission planning, including dynamic re-planning and re-tasking; and human/system integration, including situation awareness.
DARPA believes it can reduce J-UCAS life-cycle costs by having a single integrated software suite, rather than multiple segregated operating systems, and using an evolving internet-like model for its development. "COS is an open system. It is designed to be extensible, to accommodate new functionality and to take advantage of new ways of writing software," says Francis.
Achievings DARPA's vision for the COS will not be easy, hence the unique requirement for an integrator/broker. "We plan to put together a consortium-like arrangement led by the two primes. The integrator/broker is a third party," says Francis. "If the two primes collaborate, the integrator/broker does the housekeeping. In the primes run into areas of disagreement, it moves into the integator role."
"If Boeing and Northrop Grumman can learn how to work together seamlessly, then the integrator/broker's job will be really easy - like the Maytag repairman," says Heinz. "If we can't work together, then the integrator/broker will arbitrate and rule," says Winship. There will be an incentive for the two primes to collaborate, says Francis: "The cost of not agreeing is not getting there."
And getting there is not guaranteed. Putting DARPA back in charge is an acknowledgement that the hurdles to successful development of the J-UCAS remain enormous. As the concept has evolved from the relatively inexpensive and "attritable" UTA to the more expensive and survivable UCAV, so have the technological and operational challenges.
"UCAVs are design to fly at Mach 0.8-0.9 and 35,000-40,000ft - right where everyone else wants to fly," says Francis. "That makes them a forcing function for UAVs in civil airspace - they can't fly above commercial traffic like the Global Hawk can." Launching armed unmanned vehicles on long-range missions from the continental USA also poses challenges. "This is not an aircraft system, it's an air and space system," he says. "We have to lay the cable trays in space, the conduits for command and control. We have to be comfortable operating on the other side of the world."
Solutions to these issues are to be introduced incrementally during the J-UCAS operational assessment phase, scheduled to run from 2007 to 2009. The capability demonstration is structured to allow a USAF development decision in 2010, but there are several "off ramps", Francis says. "If they want an aircraft to do SEAD, strike or electronic attack, we could get a decision early on, or they could wait until the end of the operational assessment."
The ultimate goal is a system of autonomous vehicles, carrying a mix of sensors and weapons, that can collaborate as a "virtual aircraft" in a way that delivers a predictable battlefield effect, but is unpredictable to the enemy, says Francis. "If we end up with a fast-moving Global Hawk then we will not succeed at the missions we need to do."
GRAHAM WARWICK / WASHINGTON DC
Source: Flight International
