The self-driving vehicle is now a reachable asset for travelers in major American cities, after a ton of researching and testing. The Army, however, has it own need for the autonomous vehicles. To help procure the complicated technology, the Army turned to the Defense Innovation Unit (DIU) as a facilitator with folks from the industry side that can help. To learn more, Federal Drive with Tom Temin talked to David Michelson, a project manager with DIU.
David Michelson So as part of what we’re looking to do at the Defense Innovation Unit, we want to accelerate the adoption of commercial technology for the Department of Defense. And it’s no secret that private industry has invested billions and billions of dollars into the autonomous vehicle space. And there’s a lot of technology and some really innovative ideas out there that the department could benefit from. So as we were talking to a bunch of our [Department of Defense (DoD)] partners, we ended up making connections with the Robotic Combat Vehicle Office out of Detroit Arsenal. We learned that they had a pretty compelling problem on the software side for their efforts and that we could help out and really open up the aperture of the vendors that could compete. Get some new companies, some compelling, nontraditional defense companies and some first timers really, that haven’t done business with the Department of Defense before to come in and say, hey, we have something that’s compelling, unique, innovative, and we think that we could solve a big problem for the Army. So those conversations happened for quite a while as we strategize, thought about it. That’s really where the match came in. And I think that DIU’s unique position in the ecosystem and connections to the commercial space or something that the Army was excited to leverage and we’re proud to partner with them.
Eric White Yeah, this is really DIU’s bread and butter. And like you said, this is kind of what you all were created for, was helping to facilitate processes like this. So what was the steps that you all took and what role did you play in getting the Army in touch with these different companies?
David Michelson The process that we use is laid out pretty well on our website at DIU.mil. For anyone that’s interested in taking a deeper dive into it, definitely go to DIU.mil there’s some great open solicitations on there right now, and it talks vendors through how to get into submitting in a solution brief for a different solicitation. So the commercial solutions opening is the process that we use for our other transactions authorities to be able to execute contracting in a competitive process for the DoD partner that we’re working with. And that’s the mechanism that we used with this. So we really took the variety of requirements that the Army had. We talked to their user representatives, we talked to the different stakeholders, and we work hand in hand with the project manager, and the product managers and the army for the robotic combat vehicle to say, what is the problem that we’re actually trying to solve here? And we distilled that down into what we call an area of interest statement. And that’s what vendors see when you go on DIU.mil, when you see a work with us button and has open solicitations, that’s what you see. That’s what everyone takes a look at and it describes at a high level what the problem is. What we want to do is demilitarize the language that goes out to industry, to make it practical, to make it so that we can see innovative solutions. We don’t want to prescribe a solution necessarily. We want to describe what the problem is and see what’s out there that can solve it. And the barrier to entry for companies to submit is pretty low. It’s either a five page white paper or 15 slides. That’s all they have to do. They do have to answer the questions in the solicitation. They do have to address it. Like I said, that’s all explained pretty well at DIU.mil. But that gives a wide variety of vendors the chance to take a dip into the ecosystem and to see if they have something that’s really compelling for the government. And so that was the process that we worked on with them. I think it was about almost a year ago now that we launched this initial effort online.
Eric White And so you received a healthy amount of responses and got it down to five vendors. What can you tell me about the decision making process? And how did you aid Army officials in that step?
David Michelson It’s a competitive process and there’s a lot of diligence that goes into that. And the process goes through several phases where we hear very detailed proposals. We have conversations with the vendors and we really take a look at the breadth of their solution compared to the problem statement that the government’s trying to get after. So once we get through all of that and we collaborate with all of the different stakeholder, and the vendors are part of that, they’re included. Because again, we’re really trying to leverage their commercial capabilities and who knows it best they do. And so we want to make sure that we’re doing something that’s achievable and that we’ve got everyone lined up for success. So that’s really where it comes in. We bring in a lot of expertise, not just from inside the IU, but from the rest of the government and the army to really help us figure out exactly what we want. And we try to remain as flexible as we can so that we can learn along the way and address those learnings and make improvements every step and every phase of a project’s lifecycle.
Eric White All right. I think we’ve healthily satisfied the contracting part of this that we had to get through. Now lets talk some cool tech stuff. How far along really is this technology? I mean, I was just reading yesterday about how autonomous vehicles are being used right now in West Coast cities and not really a lot of people are talking about it. So what is the Army looking to do to utilize unmanned vehicles?
David Michelson There’s a whole pipeline of technologies that go into that. So if you go talk to any of the companies that are putting vehicles on roadways right now, like one of the partners that we work with on this effort, Kodiak Robotics, who has semi-trucks on the highways in Texas and other states in the South right now, you’ll hear about how it goes from an idea through the engineering process, through modeling and simulations, which we have another partner working on applied intuition, and then how we refine the models to make all of that work and exist cohesively so that, again, the Army could take a look at this and verify that, hey, this is achieving the standards that we need. We feel comfortable with the tech that we have. And so we’re still fairly early in building that full pipeline out for the Army. But we’ve had demonstrations. We made a couple posts online where you can see we’ve had several of our vendors in addition to the ones I’ve mentioned, robotics research.ai, systems as well come out and demonstrate their technology in military environments so that we can get a better grasp of where they’re at. And I’m super optimistic. I think there’s a ton of utility in this, and there’s a lot of different use cases from logistics to reconnaissance that the Army can really benefit from. And really the whole DoD combined. It’s more really trying to look at as those hard things the dole dirty and dangerous tasks that people that troops are doing now in vehicles, and how can we get them out of those vehicles so they can do things that are more complex and harder for a machine to do that you really need a person to go do. But if you’re having something move logistics from point A to point B, commercial vendors are doing that right now and they’re doing it and they’re scaling. And it’s really exciting to see that translate into a military environment.
Eric White We’re speaking with David Michelson. He’s a project manager with the Defense Innovation Unit. And so what specialty is you mentioned a little bit about what the use cases are as the Army’s trying to use unmanned vehicles. Obviously, you put an unmanned vehicle on a highway in the U.S. safety is of the utmost concern, but I’m just curious about what stipulations the Army has for their needs. If you’re sending an unmanned vehicle across a battlefield or something, earth, the conflict area, what are the specialty required that may not be needed on the home front?
David Michelson Yeah, that’s an interesting question. And the way I personally like to think of it is that when an autonomous vehicle company puts their vehicle on the interstate in the United States, they don’t want to cause an accident either. Safety is paramount. And it’s honestly, I think, a more congested and just as unpredictable environment as you would see in a military operation. There’s still clutter. There’s still people cutting vehicles off or doing wild things that we’ve all seen on the highway in the machines have to be able to react to that and respond to it. I think one of the interesting things that we’re looking at as we scale this is the ability to do what we call tele operations or remotely operate the vehicle. When there’s something that comes up that we just haven’t seen before. And that’s common. That happens all the time for all the vehicles that you’re mentioning in West Coast cities and other places throughout the United States. Like every day, I’m sure these developers are seeing things that they didn’t think up. And so they build to it and they react to it and they adjust their models and go redeploy their systems to see if they learn appropriately. And we’re doing the same thing in this project, and we’re really trying to lean on that ability for a human to come in on the loop, to adjust for an obstacle or something that they hadn’t encountered before. That helps us really kind of scale this and make it quicker for adoption I think, and safety.
Eric White As someone who has gotten a quick snapshot of this technology and well, I’m going to ask you to put your fortuneteller hat on and tell me are we going to see convoys of unmanned Army vehicles going into conflict zones or is this just going to stay as sort of an a logistical aspect for the time being?
David Michelson I think so. I think there’s definitely a desire to see those convoys and to really build as at scale so that you can accelerate operations, increase efficiency, and really give commanders in the military a lot more flexibility so that they can use their people again for more complex tasks and things that are aren’t dull, dirty or dangerous. They can go do other things. So I think that will have a huge impact. I know I can’t speak to who, but I know that there are customers out there in the department that really want this capability and we’re really hoping to accelerate that adoption. This industry is doing the same thing. They see all of the upside to this technology. And I think with the work we’re doing here at DIU and the Army, we can accelerate that into the Department of Defense.
Eric White Nobody likes driving, especially through a battlefield, so.
David Michelson Yeah, I’ve been there. I spent 14 years in the army, spent money and time on highways and roads in Iraq and Afghanistan. So if we can pull troops out of seats and increase their safety, get them away from, again, those tasks. I think this is a huge win and I think there’s a huge opportunity here for the department to leverage commercial industry and the massive amount of investment. And it’s not just the investment, but it’s also the data that’s been collected. There’s nothing like operating in the real world, and commercial industry has been doing this through several pilot programs across the country and they’ve collected tremendous amounts of data, and that’s only a positive that the department can benefit from in partnerships like these.