Teams led by General Atomics and Dynetics beat competitors at Lockheed Martin and Composite Engineering to enter the second phase of the Defense Advanced Research Projects Agency's Gremlins project, and will continue competing over the next year to build a system capable of launching and recovering unmanned aerial vehicles with a C-130.
Gremlins would allow the Air Force to eventually field multiple disposable UAVs equipped with different payloads that could fly ahead of a host aircraft on strike, intelligence, surveillance, reconnaissance or other missions.
Company officials outlined their approaches going into phase two of the program in recent phone interviews following DARPA's March 15 announcement that the two teams would advance.
General Atomics sees Gremlins as a launchpad for a small UAS portfolio it will unveil over the next three years. The company plans to add recoverable, airborne UAS to its MQ-9 Reaper and stealthier Avenger as well as across the Air Force inventory and Navy platforms like the P-8 Poseidon and future MQ-25.
"GA's position for phase one and moving forward is to create a product within the Gremlins program that's transportable to as many host platforms as possible to include things like C-17s, C-130s, unmanned aircraft," Mike Atwood, director of General Atomics' advanced programs group, told ITAF March 22. "It's not necessarily to focus at the cargo bay or the wing or different aspects of the C-130, but to focus on a system that's transportable to different aircraft and different recovery stations in the future."
Tim Keeter, Dynetics' deputy program manager and chief engineer for Gremlins, said in a March 16 phone interview the UAVs could viably launch from and return to inside the aircraft, do the same from the wing, or launch from one and dock on the other.
"In one approach, you could launch it from a pylon and recover it on a pylon," he said. "You could launch it from inside an aircraft and recover it inside the aircraft. You have to look at automation and manpower and the time of recovery, and all those factors play into different approaches."
A Lockheed Martin spokeswoman confirmed March 20 the company's C-130 would be able to carry UAVs on its outer wing pylons, depending on their size, shape and weight. The Gremlins design has to account for how large, heavy and powerful UAVs are when carrying various payloads, to figure out the physics needed to fly on and off of a C-130.
The air flow around a large cargo aircraft adds another layer of complexity, Keeter said, particularly when flying in difficult weather conditions and at various altitudes. Those environmental factors could make it easier or harder to launch or recover a UAV from the wing compared to inside the aircraft. General Atomics officials did not offer details but said their solution allows for safe, robust recovery.
Air Force Special Operations Command missions are the target use for Gremlins' initial operational rollout, Atwood said. He said whether the small UAS launch from and return to the C-130 wing or cargo bay depends on how warfighters envision their use.
"Accomplishing this objective will eventually equip the military with the flexibility to complement current mission objectives by doing things like improving standoff of manned aircraft, multiplying efforts to geolocate targets, and extending strike capability of current combat platforms," Keeter said. "It will also provide a necessary architecture for future U.S. military objectives in the area of unmanned autonomous distributed capabilities."
DARPA wants the Gremlins concept to use swarms of low-cost, attritable UAVs that can be recycled about 20 times, with less than 24 hours between missions. The more UAVs that are part of the Gremlins system, the more affordable the program becomes, Keeter suggested. Atwood added General Atomics generally believes volleys consist of more than 16 UAVs in a swarm.
The UAVs are expected to fly 300 to 500 nautical miles away from the C-130 and loiter. Keeter imagines the radius will increase over time.
During the program's 10-month-long first phase, companies developed detailed concepts of operations, system-level requirements, performance and affordability studies, and plans for phases two and three with cost, schedule and performance estimates. In phase two, which will run over the next year, and phase three, which will last about 18 months, the teams will mature systems that can help UAVs overcome the challenges of airflow, speed, proximity to the C-130 and more to successfully launch and return the small aircraft.
"When you have an unmanned system and a manned system flying in close formation, there's a lot of safety concerns because you have to close the distance," Keeter said. "You want to make sure that at all times . . . the potential for that to strike the manned aircraft is very low. There's just a lot of technical complexities that go with that type of precision and those types of safety features -- mechanical, electrical, software, you name it."
Dynetics' Program Manager Mark Miller added during the March 16 interview the challenge lies in controlling multiple Gremlins while they wait to board the aircraft and designing the means to pull them in. Engineers need to make sure the aircraft align properly without bumping into each other and are recovered quickly enough to make the idea feasible.
While DARPA's directive focuses on the C-130, Dynetics did not elaborate on how designs may differ across the aircraft's variants. General Atomics' design would not change under different circumstances, Atwood said.
"The solution that GA's proposed and is executing on the phase two is not unique depending on the mission set, the type of configuration of the air vehicle or the host system platform," Atwood said. "You can perform the same recovery mechanisms whether it be EO/IR sensors, electronic warfare sensors, kinetic capability … with the same investment."
That mindset applies to their unmanned aircraft plans as well: UAS could fly on the Reaper's seven hard points on the wing and body, or the Avenger's hard points and inside the internal weapons bay.
"If you're DARPA, you want a value proposition," Atwood said. "If GA is able to propose a system that is relevant, not only to our own unmanned platforms, but as well as P-8s and C-17s and B-1s, the larger community of aircraft that we have, you could see why it would be easy to select a company like General Atomics that's giving them a much more pervasive solution than something that's solely about a demonstration on the C-130."
Keeter added Dynetics' subsystem technology could also fly with manned-manned, manned-unmanned or unmanned-unmanned teams.
Dynetics prefers a completely autonomous system, though the Gremlins could be flown by someone on the C-130 or on the ground.
"There's just a number of different approaches that you could take in terms of when you hand off control of the vehicle and let it operate autonomously, versus when you have a man in the loop," Keeter said. "We're going to want to take the most reliable route with the technology that we can mature and implement between now and phase three."
Chris Pehrson, General Atomics' vice president of strategic development for the Defense Department, told ITAF March 22 their Gremlins technology will be similar to the company's automatic takeoff and landing capability, using high levels of autonomy.
Col. Brandon Baker, chief of the Air Force's remotely piloted aircraft capabilities division, told reporters last October the Air Force and DARPA signed a memorandum of understanding so the service's intelligence officers can explore the possibilities of command-and-control networks in relation to Gremlins. Pehrson said they regularly meet with Baker, Air Force ISR Capabilities Director Gen. John Rauch and Deputy Chief of Staff for ISR Lt. Gen. VeraLinn Jamieson to discuss key enabling network technologies and concepts of operations that could apply across the new class of small UAS.
Stakeholders are studying which datalinks would be most robust and resilient for a swarm, Pehrson added.
Gremlins could eventually serve as a testbed for payloads, datalinks, communication techniques and collaborative autonomy, Atwood said -- "kind of a flying garage shop to the AFRL community."
"The entire community's collaborating," Atwood said, noting the "early buy-in of the operations community and the futures community of wanting to be stakeholders on the ground floor and not waiting for a development program to happen and then figure out how to use it."
"I think it embodies where the defense industry's heading with more stakeholders early in the process, and to see a large organization like the Air Force want to partner with someone like DARPA, I think shows the health of the aerospace industry," he added.
The Dynetics officials said the MOU has not spurred any differences in their work.
DARPA awarded each company design contracts worth up to $21 million for phase two, according to Dynetics. One company will win a contract in early 2018 for phase three and conduct final flight tests in 2019. The Interior Department also awarded Dynetics and General Atomics "phase II, part B" test contracts on March 13 for nearly $500,000, and on March 8 for almost $1.5 million, respectively.
"Due to the rapid nature of the program and the fact that it is a demonstration (as opposed to a formal program of record), DARPA is acting as the airworthiness authority for all flight tests and funding those activities under the Part B contract," Dynetics spokeswoman Kristina Hendrix wrote in a March 23 email. "Differences in how the total amount is split between Parts A and B likely have to do with each contractor's airworthiness activities and rates."
DARPA did not respond to further questions by press time (March 23).