Space is not so much hard as it is expensive. Satellites today are expensive machines, expensively built and expensive to launch, with the understanding that, once on orbit, they can work for years.
That calculus assumes several eggs in every pricey basket, and as space moves from a home for military satellites to a domain where nations prepare for actual combat, building resilience in orbit means rethinking how satellites are done. It means rethinking costs in the billions and imagining them instead in the millions. And to the Defense Advanced Research Projects Agency’s Paul “Rusty” Thomas it means creating a whole new ecosystem for payloads and launches.
Thomas is the program manager for Blackjack, a DARPA initiative that wants to pilot a constellation of cheaper satellites for military communication, with the costs low, uplinks up and the resilience of the whole constellation baked-in.
C4ISRNET’s Kelsey Atherton spoke with Thomas about the program.
C4ISRNET: There’s a lot of interest in both low Earth orbit [LEO] and constellations of satellites. What is DARPA’s specific goal with Blackjack?
PAUL “RUSTY” THOMAS: Blackjack, as an architecture demonstration, will build a portion of a constellation, looking at about 20 percent of a fully proliferated LEO constellation. That’s a range of 20 satellites, 20 percent of the 90 to 100 satellite constellation, which would give a ground user three to four hours per day or more of theater-level operations so that we could actually demonstrate what we’re going to do with a full, fully proliferated 24/7 constellation that covers the entire Earth and gives global constant coverage and global constant custody.
C4ISRNET: What was the logic behind accepting separate proposals for busses and payloads?
THOMAS: Most exquisite spacecraft we built have been married to the bus and payload from Day 1. That’s a wonderful model for exquisite spacecraft. But we’re trying to build a proliferated LEO payload ecosystem — like the commercial commoditized bus ecosystem — that can match the numerous types of payloads. To do that you don’t want to just show that one payload matches great and then move forward. That just gives you a great payload.
To try and build that ecosystem out, you want to go to at least Program Design Review with the payload developers working to a generalized initial design covering numerous types of commoditized busses. Once you get deeper into the design phase, match that payload to a bus, which allows a large range of payloads to be developed.
C4ISRNET: There’s a lot of commercial interest in this space; does that pose any risk to deploying a new constellation?
THOMAS: The goal of Blackjack is to prove you can leverage commercial approaches with potentially lower costs, lower cycle times, lower times for design and build. It also comes with the issue that we’re not directing the approach to building the bus, we’re not directing how the constellation is put together for these folks; therefore, the rest is getting the government itself to do that match and to put our systems into play in a way that marches in lockstep with them without directing their commercial elements will play. That brings risk. We have to learn how to do business a little different than it’s been done in the past, and to move a little quicker than the government has in the past.
C4ISRNET: So, there’s no risk of LEO being too crowded to accommodate more constellations?
THOMAS: No. Well, I wouldn’t say no risk, there’s always risk, the mega constellations that you’re starting to see FCC filings for look like they’re going to put hundreds, and some of them into the 10,000-plus range, and that’s certainly going to be a challenge and it’s going to be a risk.
Fortunately, we have air traffic control systems on the ground that cover large numbers of aircraft in the air at any given time. We haven’t actually taken that step into how to manage large numbers of spacecraft in space yet, but we believe that all the technology is there and it’s just a matter of implementing an area where the government is going to be tracking what the commercial folks are doing.
There’s a risk — it’s not major, space is big — but you absolutely need to track the spacecraft and make sure they can deorbit. But in terms of putting thousands or even tens of thousands of satellites into low Earth orbit, all of that seems very feasible and is not in the high-risk bucket.
C4ISRNET: What’s the rough timeline you’re expecting for demonstrations?
THOMAS: For the 20-satellite constellation, we plan to have the first two spacecraft that we have integrated to the commercial busses and the payload together ready by the end of 2020, with launch by early 2021. We will follow that in 2021 with the rest of the 18, once we’ve confirmed the first two are fine. We will have the full demonstration capability running late in 2021 with an expectation of theater-level autonomous operations from low Earth orbit in 2022.
C4ISRNET: One argument for satellite constellations and against exquisite satellites is resiliency. How does that work here?
THOMAS: You get a lower cost, the individual node becomes a bit expendable, you don’t build your resiliency around the individual node, you don’t try to protect that spacecraft to the nth degree like in exquisite billion-dollar-plus craft. If the Blackjack model works, spacecraft will be in the $3 million to $4 million range, $2 million to $3 million to put it into orbit. We’re talking about a $6 million node, including the cost of getting it into space. Therefore, it’s less than the cost of a high-end munition.
The constellation itself becomes your resilient element. You can put your high-level availability, reliability and mission assurance at the constellation level instead of at the node, because of the numbers you’re putting up. If one satellite has fallen, its replacement is coming over the horizon 10 to 15 minutes later. You have a different approach to resiliency, large numbers of spacecraft in play, which totally turns some of the counterspace elements on its ear.
C4ISRNET: What counter-space elements might this be especially resilient against?
THOMAS: You now have low-cost nodes, so a lot of the direct ascent type of methods out there no longer makes a lot of sense. Of course, you still have varied threats from non-kinetic and cyber. We still need to protect the constellation against all the other types of threats out there, so it probably helps the most on the kinetic side, but it certainly gives you lot of resilience in all the areas.
C4ISRNET: What kind of communications presence will this enable?
THOMAS: Blackjack is aimed at leveraging the new mesh networks being set up by these commercial companies. A user currently in the DoD might need to look up at two or three different options in space to actually talk and do communications in this space segment. Once we link up and do encryption, the user on the ground will look up and see hundreds or more potential network nodes overhead at any given point on the planet, North Pole to South Pole; it’s going to drastically change how the DoD does communication.
That is a bit independent of what Blackjack is going to do. If the commercial companies succeed and come out, that capability, call it raw gigabit-per-second class, not all of them it. But they all have many megabit data links from one point of the planet to another, at very low latency, 100-200 milliseconds, so you do really change the game for how any user, DoD included, does global communication.
C4ISRNET: Is a desired end goal of Blackjack specifically a redundant spaceborne network that can function independently if access to internet on the ground is cut off?
THOMAS: If you have a problem with your terrestrial network — whether it’s a ground network system or point-to-point comms, fiber optics or others being interfered with — the space mesh network provides the ability to move the data up, move it through the space mesh, and move it back to the ground, without any other system being involved in that data transition. The switch network that Iridium has up right now, it’s low bandwidth but a wonderful system in terms of moving data from one point to another on the planet through the Iridium gateways that DoD and its users have worldwide. Move that up to high broadband access, and not just two or three satellites overhead but dozens or hundreds, and it really does move us into a new realm.
C4ISRNET: At what point in the program do bus and payload link? Is there a point where they’re demoed together?
THOMAS: In the [broad agency announcement] out right now, you can see we’re looking for multiple payloads to go at least through phase one, potentially multiple buses to go through phase one. As we progress the programs through the preliminary design review into phase two and get critical design review, first two spacecraft built, we’ll be selecting the ones to continue deeper and deeper into the program to match up and do the demo. We’ll start with a wide range and narrow down to a smaller set to actually do the demonstration with a secondary objective of showing why a huge payload will work, why different types of payloads will be successful in this type of architecture, even though we’ve only got one or two of them.
C4ISRNET: What does the future of Blackjack look like?
THOMAS: We are looking at large numbers of types of payloads. We very much want to get into a rapid tech refresh cycle … putting up payloads every two or three years that are newer version of the ones that have gone previously, have an open architecture standard so we can update over the air with better algorithms.