Science Projects with Purpose

June 27, 2018

An Air Force Magazine illustration of a conceptual USAF air-breathing hypersonic weapon (top), utilizing its scramjet engine after separating from its solid rocket booster. Below, a Russian hypersonic missile speeds toward its target after jettisoning its exhaust fairing cap. Illustration: Mike Tsukamoto/staff

The Defense Advanced Research Projects Agency turned 60 this year, and its director, Steven H. Walker, feels he has a clear mandate from Pentagon leaders and the White House: Focus on the threat.

Testifying before Congress in March, Walker said the new National Defense and National Security S?trategies leave no doubt that the US has returned to an era of great power competition. This makes it all the more urgent for the US to stay militarily ahead of adversaries like China and Russia.

DARPA aims to keep the US military strategically surprising while avoiding being strategically surprised.

In the Fiscal 2018 budget request, DARPA asked for just over $3 billion for its wonder-weapon programs. For Fiscal 2019, it is seeking $3.4 billion.

Created in 1958 as ARPA (the acronym minus “Defense,” which was added in 1972), the agency was designed to beat the Soviet Union to the high ground of space. From that early imperative to modern achievements like tiny GPS receivers or automatic voice recognition, DARPA’s philosophy has been to turn “revolutionary concepts and even seeming impossibilities into practical capabilities,” Walker said.

Walker, with 30 years of experience at the Air Force Research Laboratory, DARPA, and the Air Force, took the reins of the agency last year.

The agency’s primary focus areas, “which I’m calling foundations,” are all directly tied to threats facing the United States, he explained in a March meeting with defense reporters. The mission areas are: (1) Defending the homeland; (2) Deterring and prevailing against peer competitors in geographically important areas of the world; (3) Effectively looking at how the US will counter insurgency and counterterrorism across the world; and, (4) Building upon DARPA’s proven strengths, he said.

The fourth “focus area for me is our foundations, and that is really what DARPA has always done—which is pay attention to where technology is leading us,” he told reporters. When the agency can successfully spot trends, it can “advance that technology to develop future capabilities for the US and national security,” Walker said.

The world, he said, is “experiencing deeply disturbing technical, economic, and geopolitical shifts” that threaten America’s “preeminence and stability,” Walker told members of the Emerging Threats and Capabilities subcommittee in his March testimony. While these are contrasted by “some remarkable and even astonishing scientific and technological advances,” Walker added, these “dueling trends of unprecedented opportunity and risk” guide DARPA’s strategic priorities.

Chief among them is hypersonics; the development of vehicles that can fly within the atmosphere at speeds in excess of five times the speed of sound. Walker told Congress in testimony that DARPA’s pursuit of hypersonics carries “a particular sense of urgency due to the rising pace of related research by peer adversaries” such as China and Russia.

In late January, Air Force Gen. Paul J. Selva, Vice Chairman of the Joint Chiefs of Staff, told defense writers in Washington, D.C., that China has the lead in hypersonics because it’s made the technology “a national program” and is spending lavishly on “solving the problems of hypersonic flight [and] hypersonic target designation.”

Two months later, Russian President Vladimir Putin touted a new missile, the “Kinzhal,” which he described as an “invincible” hypersonic weapon that possesses unlimited range.

Steven Walker, DARPA head, speaks with reporters after a roundtable discussion in Washington, D.C. Photo: Jim Garamone/DOD

Walker has called for a national effort to win the hypersonics race. Speaking to reporters weeks before his March testimony, Walker called Fiscal 2019’s budget request of a 136 percent increase for hypersonics research “a good first step,” but insisted more is required.

Funding for hypersonics jumped from $85.5 million in Fiscal 2017 to $108.6 million in the Fiscal 2018 request, leaping again in Fiscal 2019 to $256.7 million.

In late April, USAF awarded Lockheed Martin a $928 million contract to prototype its Hypersonic Conventional Strike Weapon. DARPA-specific efforts into hypersonics include the Tactical Boost-Glide Program and the Hypersonic Air-Breathing Weapon Concept. The Air Force Research Laboratory also has hypersonics-specific programs.

“We’re going to start flying these systems in 2019. You’ll see lots of flight tests,” Walker told reporters. “We’re excited that these will be systems that will be very capable,” and can be launched from “standoff” distances, meaning outside the range of adversary air defenses.

“We do need an infusion of dollars in our infrastructure to do hypersonics,” he said. “Most of our programs at DARPA are testing in one facility. … If you look at some of our peer competitors—China being one—and you look at the number of facilities they’ve built to do hypersonics, it surpasses the number we have in this country and is quickly surpassing it by two or three times.”

Walker noted, though, that “no domestic security threat today exceeds that of a nuclear or radiological ‘dirty bomb’ detonation.” Detecting the signs of nuclear testing or release can be achieved for high-emitting radiological materials, and a nuclear attack on a city is obvious. But today’s sensors are “too large and expensive” for wide deployment into urban environments where smaller radiological warfare might take place.

DARPA’s SIGMA program aims to achieve cheap, capable, and easily transported radiation detectors.

Testing of SIGMA is underway. In 2016, DARPA worked with authorities in New York and New Jersey to field 100 devices, and Washington, D.C., where they fielded 1,000 devices and tested them at “critical transportation hubs,” Walker said. In Washington, for example, DARPA installed 73 radiation and nuclear detectors in ambulances. The vehicles logged 100,000 hours of testing and succeeded in identifying “thousands of radiation sources.” These included many “innocuous ones, like natural granite,” according to DARPA, which requested $38.6 million for more research into SIGMA in the coming fiscal year.

“SIGMA detectors can readily distinguish between these kinds of benign sources and threatening ones,” the agency’s program release read. “Equally important, the SIGMA detectors provided detailed background radiation maps of the District against which future sources may be more easily detected.”

A Kromek handheld radiation isotope detector. The devices provide real-time data to detect radiation. Photo: DARPA

Another example of DARPA’s defense portfolio is its Network Defense program. Sifting through terabytes of DOD Information Network (DODIN) data, the program looks for harmful cyberspace events.

Walker said the effort brings US Cyber Command and Army Cyber Protection Team members together with its own experts; the three entities collaborating to identify and mitigate cyber threats to and in DODIN traffic. He said the program identified persistent threats within days of operating and over time detected five criminal malware infections and anomalies, all emblematic of the program’s success in the cyber realm. DARPA wants $116 million to keep running Network Defense in FY19.

The program has also created more than 40 scalable methods now used in other military, and even commercial, environments, Walker said, noting more than 60 exploitations identified in the private and military sector as a result of applying DARPA’s techniques. The program is very similar to the Air Force’s Cyberspace Defense.

DARPA is also looking to a computing future free of traditional circuits and boards.

The 2017 electronics resurgence initiative aims to upend board technology, disrupt the market, and create what Walker called “leap ahead” technology for circuits. There’s $111 million in DARPA’s FY19 budget request to cover multiple facets of this potential enterprise-shifting tech, from fostering working relationships with companies like Intel and Samsung to partnering with universities. DARPA calls this effort the Joint University Microelectronics Program.

Not all of DARPA’s technology initiatives have to do with hardware. The agency is dabbling in the growing universe of gene editing. DARPA’s Safe Genes program is aimed at protecting troops from diseases and mitigating the threat from biological warfare, and developing advanced, synthetic bio-based materials.

“The steep drop in the costs of genomic sequencing and gene-editing tool kits, along with the increasing accessibility of this technology, translates into greater opportunity to experiment with genetic modifications,” a DARPA spokesperson told Air Force Magazine. “This convergence of low cost and high availability means that applications for gene editing—both positive and negative—could arise from people or states operating outside of the traditional scientific community.”

An airman prepares to launch a small UAV at Camp Roberts, Calif. Photo: USAF

Walker said to counter such people or states requires the US to have “more complete knowledge” about how gene editors and similar technologies function. “In parallel,” he added, “demonstrating the ability to precisely control gene edits, turning them on and off under certain conditions or even reversing their effects entirely, will be paramount to the safe translation of these tools to practical applications.” DARPA is asking for $20 million for genetic technologies research in FY19.

Walker’s final example in the research frontier is also a national talking piece: Artificial Intelligence. From the White House’s announcement of a Select Committee on Artificial Intelligence (DARPA will have a seat on that panel) to USAF’s revitalized research partnership with the National Science Foundation, AI is carving out a central space in the Pentagon’s future.

DARPA is working on how machines respond to new environments and information—what Walker called “third generation,” or explainable AI.First-wave AI was about building smart systems, which allowed robots and machines to identify and predict trends by analyzing databases, along the lines of Google search results or airline ticket pricing.

Second-generation AI, Walker explained, is basic machine learning, or machine “training,” by which computers are taught to recognize a particular building, for example, so they can find one in a video feed and alert a human operator. The next step is getting the computer to explain why it identified the building, not just doing so. The computer’s rationale may be critical in then making a lethal choice about whether to fire.

Third-generation AI deals with unpredictable variables on the battlefield. What if an enemy airplane doesn’t exactly match the definition fed to the computer? What if the aircraft had some extra control surfaces? What might trip up the AI? This is where explainable AI (XAI) comes in, for which DARPA wants $22 million in Fiscal 2019.

“New machine-learning systems will have the ability to explain their rationale, characterize their strengths and weaknesses,” Walker said, “and convey an understanding of how they will behave in the future.”

DARPA wants to give ground troops a simple and unobtrusive way to access large amounts of data about their surroundings. Squad X Core Technologies anticipates a future where global stability requires Americans on the ground in foreign, hostile territory isolated from friendly networks. Squad X will protect squad members even when cut off from support and capabilities like GPS. DARPA wants $28.5 million in Fiscal 2019 for Squad X, which is a four-pronged program comprising precision engagement, nonkinetic engagement, squad sensing, and squad autonomy.

A DARPA graphic explains the Squad X concept. Graphic: DARPA

Precision engagement allows ground forces to engage a threat as far as a kilometer away, even if it’s not in sight. The component would be compatible with current weapon systems and not impose any new weight or operational burdens. Distributed guided munitions are being considered, according to DARPA.

In the second, troops could nonkinetically disrupt an enemy’s command and control, communications, or ability to use unmanned assets.

Squad sensing is the inverse of precision engagement, allowing for threat awareness up to a kilometer. There would be multisource data fusion and autonomous threat detection.

Finally, squad autonomy will provide squad members with awareness of where everyone in their team is located and where GPS has been denied. This technology involves collaboration with unmanned or ground systems and specifically involves human-machine teaming.

“The goal is to speed the development of new, lightweight, integrated systems that provide infantry squads unprecedented awareness, adaptability, and flexibility in complex environments,” Walker said.

While DARPA is frequently an agent of change for the US military, its wide portfolio can sometimes prove a hindrance, especially when paired with a plodding and outdated Pentagon acquisition process.

“Despite all our efforts,” Michael D. Griffin, undersecretary of defense for research and engineering, told Congress in late April, “we are constantly challenged to maintain science and technology superiority.”

Still, the agency’s successes over time are undeniable, so much so that other nations and even other portions of the US government often try (and fail) to emulate DARPA’s successes. “I’ve actually over the last couple of years met with many countries—South Korea, Japan, UK, France, Germany, to try and help them understand the DARPA model and how it could work for them,” Walker told defense reporters in March. “Every culture’s different, though, and one of the reasons why it worked here is, compared to some other countries, we have a risk-taking culture. We have folks who are willing to come to a place, have a job for three or four years, and then get booted out. Which is actually what happens.”

Other nations don’t necessarily have the professional mobility that is common in the United States. “People generally know they’re going to go from DARPA and get a good job somewhere else,” Walker noted, so “we get new people in the door with new ideas all the time. We’re not bound by 10-, 20-, 30-year-old thinking, which can happen in some places. And it adds to the innovative culture of the place.”

The other reason DARPA is successful, Walker continued, is the the agency has buy-in on its model across the government. “They give us a lot of freedom. … To make decisions, to think differently, and to start and stop our own programs,” he told reporters.

“We get very little top-down direction, which I think is [important] if you want an organization to continue to produce out-of-the-box ideas, projects, to continue to disrupt the status quo and question, and you want that organization to be free to have some autonomy and some flexibility,” he concluded.