DARPA has no doubt stimulated many important innovations over the past 60 plus years, from the internet and GPS to drones and autonomous vehicles. But it is a well-known problem that there has been a “valley of death” between the technology labs and the traditional acquisition system. The number of successfully transferred technologies has been less than desired. The purpose here is to discuss how, on the margins, DARPA can be made more effective at delivering value to the military force structure.
DARPA is funded primarily from budget activities 6.1 to 6.3 in the RDT&E appropriation. That ranges from Basic Research to Advanced Development (such as experiments). Generally, BA 6.4 funding is used for prototype competitions. For almost all major programs, there is either no prototype competition (e.g., Columbia-class submarine and the B-21 bomber), or they are handled by a dedicated program office (e.g., the JLTV drive-off and the JSF fly-off). So many major programs have in effect already been “transitioned” without substantial input from DARPA.
DARPA struggles to transition its work from 6.1-6.3 into 6.4 (which is where a prototype competition would start, if there is any). That transition coincides with organizational transitions, from the labs to the program offices.
So it appears that because of challenges in bridging the “valley of death” to the program offices, DARPA has decided to focus inwardly on Basic Research. It stimulates innovations that are simply way too early for use in DOD operations within 5 or 10 years or many more years. Its not that they are not important, and may not pay off militarily at some point. But their projects are too early, and in the market, being too early is as good as being wrong. They aren’t creating user-friendly applications in the defense world, as it were. They are making forays into possible types of future infrastructures.
DARPA seems to thrive on basic research, which, by definition, cannot be particular to military applications. Here is one example that sounds typical of the incentives and rewards provided to DARPA:
DARPA was largely responsible for launching the fields of semiconductor spintronics and quantum spintronics, the latter of which has also helped drive the emerging area of quantum computing … This was high-risk science and, at the time, not obvious that the underlying physics and engineering would work out so well.
That started in the 1990s. We’ll see when benefits accrue to defense. Here’s another recent example:
The Molecular Informatics program brings together a collaborative interdisciplinary community to explore completely new approaches to store and process information with molecules. Chemistry offers an untapped, rich palette of molecular diversity that may yield a vast design space to enable dense data representations and highly versatile computing concepts outside of traditional digital, logic-based approaches.
Again, interesting and important, but way too far ahead. For this next one, DARPA is saying explicitly that there are no military applications:
The Pentagon is studying whether insects can be enlisted to combat crop loss during agricultural emergencies… DARPA’s program manager for Insect Allies, Blake Bextine, pushed back on the Science article, saying the program is solely for peaceful purposes.
DARPA’s famed Grand Challenge in 2006 seems to have stimulated commercial market advancement in driverless vehicles first, which in turn boomeranged back to the traditional defense acquisition process years later.
My impression from the very excellent Voices from DARPA podcast series is that many project’s benefits are expected to accrue to the soldier in terms of decades.
Of course, these are generalizations and DARPA has transitioned important projects. I am a fan of the RSGS satellite repair and upgrade robot, though the project has encountered difficulties with the contractor.
Now, I believe that the government should, of course, fund basic research. Many of the things DARPA seeks to achieve, it seems to me (as an outsider), are breakthroughs that have civilian and military uses. Indeed for most basic research, one can imagine a wide range of applications. But with so much advance in so many sectors, DARPA would impact the DOD a great deal more if it focused more on integrating research into defense systems through rapid and competitive experiments (if not prototypes). In other words, there is a major gap in the DOD for doing combinatorial innvoations. Putting different components together and trying designs out to learn in a rapid and iterative way. On the margin, it could do more benefit by emphasizing BA 6.3 or even 6.4 more than BA 6.1.
Now, DARPA seems to have developed a very productive culture closer to the basic research side of things. It would be a shame to throw the baby out with the bath water. I’m not saying DARPA does no basic research. I’m not saying DARPA should only focus on prototype competitions for systems that the military can try out and industry can produce in short time frames. But a shift in the direction of rapid and austere prototyping with user feedback, I believe, would be beneficial.
Leave a Reply