Podcast: Skyborg Express

Skyborg could also be a cargo delivery aircraft. The program launched by Air Force Research Laboratory is mainly known for developing artificial intelligence for a new family of unmanned combat air vehicles, but the program also includes Volansi's cargo delivery system.

We talk with Volansi CEO Hannan Parvizian and newly appointed board member Will Roper, formerly assistant secretary of the Air Force for Acquisition, Technology and Logistics, about how Volansi could become the next Skyborg. 

Don't miss a single episode. Subscribe to Aviation Week's Check 6 podcast in iTunes, StitcherSpotify and Google Play. Please leave us a review.

Below is a rush transcript of this episode.

 

Steve Trimble:

Hello. Welcome to the Aviation Week Check 6 Podcast. My name is Steve Trimble, I'm the defense editor at Aviation Week. And today on your host for what I think will be a very interesting discussion. Something very interesting happened just before Christmas last year. On December 23rd, the air force awarded a contract vehicle within the Skyborg program with a notional value up to $400 million to VOLY Defense Solutions, a subsidiary of a start-up company named Volansi. Until that point, we had heard of Skyborg mainly described as a concept for developing an autonomous control system that could be inserted into a new family of relatively cheap unmanned aircraft systems, which would help them team up with a combat aircraft like fighters and bombers. By selecting Volansi however, the air force expanded this Skyborg concept to include autonomous cargo delivery systems for the first time. Volansi is in fact, one of the pioneers of the burgeoning market of autonomous cargo delivery aircraft.

            Right now, the Voly C10 vertical takeoff and landing system is flying in a pilot program in North Carolina. The electric powered C10 picks up about 10 pounds of cold chain medicines from a factory owned by the Merck pharmaceutical company in Wilson, North Carolina, and transports that precious cargo to a local hospital center about two miles away. So it's a good time to find out more about Volansi and how it's delivery drone can be applied to the military through Skyborg or other channels. And we've got a great panel here today to make that happen. I'm joined by Volansi CEO, Hannan Parvizian, Aviation Week senior technology editor, Graham Warwick, and Volansi's newly appointed board member Will Roper, whose name is undoubtedly familiar to many listeners of Check 6 as he is the former assistant secretary of the air force for acquisition technology and logistics until about two months ago. Good morning gentlemen.

            [crosstalk 00:02:09] Well, I do want to talk about Skyborg and the military applications. But first, let's just get a better sense of how Volansi got started and its path into the commercial market to this point. So Hannan, I understand you at one point you were at Tesla and this idea came to you at Tesla. So how did that happen?

Hannan Parvizian:

Yeah, thanks Steve for having me. The idea for Volansi started over five and a half years ago now, when at Tesla we had issues shipping parts and supplies either to our assembly line or our factory to our service centers. And that resulted in a lot of downtime costs for the business and also in core customer satisfaction. So the idea from Volansi was to create a delivery service for enterprise customers like Tesla deliver these critically needed supplies as fast as possible using autonomous aerial delivery vehicles or drones.

Steve Trimble:

Well, so how far has that progressed and are you doing that for Tesla or are you planning to do it for Tesla?

Hannan Parvizian:

Absolutely. I would love to do it for Tesla. I mean, they're the original routes that I had envisioned was from our Fremont factory or Tesla's Fremont factory to Reno Gigafactory. So that was where I thought there would be a lot of traffic back and forth. We're still ways away from doing that, not from a technology perspective, but more from a regulations perspective here in the US. However, we've taken that same idea and expanded it to other customers and other markets. And mainly we operate with enterprise customers and the oil and gas and mining industries right now outside the U S. Predominantly in emerging markets like Western Africa, we service medical customers, as you mentioned, doing cold chain logistics of vaccines and medical supplies, both here in the US and outside the US. And for the past three years, we've been servicing military customers, delivering medical supplies and other critically needed supplies that are needed in the battlefield or on a base.

Steve Trimble:

So how did the channel open up to the Skyborg Program for you at Volansi?

Hannan Parvizian:

The Skyborg Program was interesting to us because it was combining AI and autonomy with a new generation of unmanned aerial vehicles. When we looked at the program, we didn't want to participate in building the vehicles that were for high-speed missions, fight along [with] jet fighters, but our angle was how could we support the six mission sets that they had defined there would are the tall or vertical takeoff and landing aircraft combining our autonomy and software and servicing the field with critical supply or other central operations that are needed. And just to give you some background, what Volansi specializes in as we provide services to customers that have payload requirements of 10 to 200 pounds and the vehicles have a range of 50 to 500 miles for an endurance of up to 30 hours on a larger platform. So that gives us a lot of versatility and flexibility in the mission sets that we could accomplish with our family of vehicles.

Steve Trimble:

Well, so Dr. Roper, I understand the idea of this Skyborg Program started out at the Strategic Capabilities Office with the Avatar program, and then moved to AFRL after you became assistant secretary. Take me through that. What was your original plan and conception for what this program was and how does Volansi fit into that?

Will Roper:

Steve that's taken me way back to the Strategic Capabilities Office. And during those times, I just wanted to see artificially intelligence supported drones enter the military in a way that could work with existing manned aircraft. That was the origin of the Avatar program that has now morphed into the loyal wingman concept and Skyborg. And I still think that's a potent concept for the battlefield. But as Hannan points out, there are a lot of applications of manned-unmanned teaming, and AI is going to be ubiquitous on the future battlefield, including doing logistics and support in critical operations. And so I view Volansi as well poised to bring in a different form of this man unmanned teaming concept, so that we do time critical logistics faster on the battlefield or during disaster relief or during rescue. And by doing that within the military, we're actually de-risking this technology for broader commercial applications.

            If the military trust it for delivery in most critical missions under duress where safety and mission are on the line, then that's going to convey a lot of confidence in commercial markets I expect. And I'm a believer that we're going to have manned-unmanned teaming for the foreseeable future on the battlefield. I think that people do things very well. AI do different things well. Both have their vulnerabilities, both have their strengths and a teaming relationship is going to be the next ethic of warfare and programs like Skyborg or hurrying that epic along. And I want to see the US and our allies and partners continue to lead on the forefront of the battlefield, including this very interesting area of technology.

Steve Trimble:

So can you delineate for our audience Dr. Roper, what they mean when they talk about an autonomous capability, like the RQ-4 Global Hawk versus an artificial intelligence control system like Skyborg and what the difference in capability is that we're talking about?

Will Roper:

The difference is going to be whether the algorithms that support that vehicle are learning on the battlefield. So in the case of something like a Global Hawk, it's a capable system, but the way it's going to perform is deterministic. Inputs will determine outputs. As we bring AI and machine learning into the battlefield, there will of course be a training set that we will have used to prime those algorithms to be supportable on the battlefield. But systems are going to evolve as they experience new data, new threats, new friendlies, and then make decisions because of that, and are then reinforced by operators that say, yes, no, get better or get out. So I think the difference is going to be that the AI systems are going to feel more like operators on the battlefield. When I got artificial intelligence into the U-2, right before leaving the air force, the air force chose to treat the AI copilot as a pilot, not as another piece of war fighting equipment, because the AI copilot has to be trained.

            That training has to be up kept just like a human would. So when you bring AI into the battlefield, you need to think of it more as an operator. Is their training relevant? Is it prepared for this mission? And if it is, it will have greater adaptability than a set of closed form algorithms will have, but it's going to push an entirely different epic of readiness for the military, one that has both people and machines as part of the readiness equation. And the time to start writing that playbook is now before our adversaries right at ahead of us.

Steve Trimble:

But let me pass it over to Graham. And then I might have a question at the very end, but go ahead Graham.

Graham Warwick:

So my interest is starting on the commercial side, we see a lot of demonstrations. They've been given a boost by the pandemic and the delivery of medical related payloads, but a lot of them are just demonstrations, particularly here in the US. I'm interested as to how close you think we are to leaving a viable sustainable commercial delivery capability in place, or are we still missing some of the pieces to be able to leave in an ongoing commercial operation.

Hannan Parvizian:

Well I would actually push back on that. I wouldn't say all of it is demonstrations. What we are doing in North Carolina and what other companies have been doing both in North Carolina and other states are ongoing sustained operations. So there are daily deliveries multiple times a day with real products delivered to real patients or real customers. I think where we're still falling short is scaling that and making it ubiquitous around the country. And I think the path to that is going to be working with the FAA, making sure all of the companies that are going to be in this delivery business get their Part 145 certifications, get their types certifications, and then integrate into the national airspace through UTM systems and other onboard technology that are required for automatic traffic and collision avoidance. So the path to that is clear in many senses. And I think [it’s] going to shake out over the next two to three years. And by 2024, I feel confident that we'll see millions of drones in the air doing deliveries and also other missions and at large.

Graham Warwick:

Interesting. And also I have to look on the military side, again this is something the military has been playing with for decades. We've had lots of tests and demonstrates as many of them to the outside world, looking very duplicative. They're all sort of different agencies looking at essentially the same thing. You say that you're providing services today for military customers. So how close are we to actually getting routine unmanned logistics into the field?

Hannan Parvizian:

I would say similarly that as we started with these programs a couple of years ago through the Defense Innovation Unit and got to start working with the Navy Medical Service and have expanded since then, I would say similarly, we're going to see a large adoption curve or the toll logistics or unmanned logistics, both aerial and on the ground and a battlefield over next coming years, especially if we can provide it as a service to DOD customers, rather than have to sell and the capability or solution set that they have to learn and figure out how to operate these vehicles themselves. So then similarly, we're going to see they large adoption of this technology over the next couple of years in the battlefield.

Graham Warwick:

So Dr. Roper, do you feel that we have been playing with this idea for a long time? Do you think we are about to break through and deliver useful capability?

Will Roper:

I do. And this is also satisfying for me because one of the programs that I didn't pull the trigger on at the Strategic Capabilities Office was a last mile logistics program. And the electric vertical takeoff and landing technology has come so far since five years ago when I first looked at it. It's been aided by a lot of advancements by the automotive industry. Technology is continues to be reviewed as safe and ready to operate by the military and other officials. And I think we're going to see it transition into military operations soon. But I have to hit Hannan's point that providing it as a service is a really important factor for the military. The use cases are out in operations and maintenance commands those that don't have research development test and evaluation money. So the military procuring hundreds, thousands of these drones to do last mile delivery is not going to be a good acquisition model.

            Having a company provide and facilitate the delivery, whether it's through an hourly or a licensing schema is going to be much better for those operational commands who procure a lot of services today as it is. It's going to fit into their existing model where they can dial up and dial down based on need. So the two things that are coming together are the vital technology that has been driven by advances in batteries and many other technologies from the automotive industry plus companies that are going after a commercial use case like Volansi that are pivoting that into the military in a way that will be easier to purchase.

            And the only thing holding them back is getting a military flight release, which is what Skyborg is meant to help facilitate. So they can walk into that operational command and say I've got approval to fly in military airspace. I have a service model that fits inside your color of money, and here are the rates and here's what I give you. And in the case of Volansi, I think it's great that commands will see lower prices for many of their missions, but with higher responsiveness. And that's going to translate into saving lives, saving missions, saving time. So there's not really a downside to it. So I agree with Hannan the next two years or the year I expect the inflection point will be passed and the adoption curve will start going up.

Steve Trimble:

With that, I think we can wrap up this Check 6 Podcast interview. I want to thank Hannan Parvizian and of course Dr. Will Roper for attending and Graham Warwick and our producer guy, Ernie Howe in London for putting all this together. And that's it. So you can download our podcast on a variety of different channels. Please leave a review at the end and thank you very much for listening

 

Steve Trimble

Steve covers military aviation, missiles and space for the Aviation Week Network, based in Washington DC.

Graham Warwick

Graham leads Aviation Week's coverage of technology, focusing on engineering and technology across the aerospace industry, with a special focus on identifying technologies of strategic importance to aviation, aerospace and defense.