PARIS — A start-up company with financial backing from Paul Allen’s Vulcan Capital among others will begin launching a constellation of 60 50-kilogram satellites providing one-meter-resolution optical imaging this year.
Seattle-based BlackSky Global is using Spaceflight Services to build the satellites and hunt for launch opportunities, with optical imagers provided by Harris Corp.’s newly acquired Exelis divsion.
One-meter imaging is becoming common, but BlackSky Global Chief Technology Officer Peter M. Wegner said the company is banking on “persistence,” or a rapid revisit time over a given area, to build the business case for the constellation.
Operating in a 450-kilometer orbit, each satellite will have a lifetime of about three years, a business model that Wegner said gives the company time to perform technology refresh to update the constellation.
The downside is that BlackSky must be in a continual capital-expenditure cycle — OK if satellite costs come down but more challenging with respect to finding launch services. Here, too, Wegner said the market is changing fast enough that relatively inexpensive launches should be available.
BlackSky’s first launch, of two satellites, is scheduled for December, as part of a Spaceflight Services’s Sherpa Tug carrying dozens of small satellites to low Earth orbit aboard a SpaceX Falcon 9 rocket.
Wegner, a former director of the U.S. Air Force Operationally Responsive Space Office at Kirkland Air Force Base, New Mexico, outlined BlackSky’s business plan in an interview.
Q: Who are your financial backers?
A: Investors include RRE Venture Capital and Paul Allen’s Vulcan Capital, with additional investment from Razor’s Edge Ventures.
Q: Are you looking for financing to close the business case for a 60-satellite constellation or are your investors committed to taking you through to that point?
A: I can’t comment on that. The business plan includes the capital to build the full 60-ball constellation. We are working the financing for that with our board of directors, which includes all our existing financial backers. There are a number of different instruments the board is considering to finance the whole constellation. The mix of different financing instruments hasn’t been finalized yet.
Q: Can you provide a ballpark sum for the cost of building and launching the satellites?
A: I can’t at this point. We are sensitive about that because we think it’s one thing that separates us from what the others in the market are doing. We’ve worked really hard on that number to be competitive.
It’s not a hard number, it’s just an estimate. Depending on market demand it could be more than that or fewer than that. We view this business model as an imaging services business. It’s space as a service. We’re all familiar with transponder leasing as a service business in satellite telecommunications, and we are doing the same thing. We are leasing out capacity to the customers.
Q: Four more satellites are to launch in 2016?
A: We’re in development on the next block of satellites now. We can produce on a regular cycle and allow us to upgrade our system, so we are already on Block 2 to launch next year. We are negotiating launches for those. We don’t have launches finalized.
Q: How many of these Spaceflight-built satellites do you need in orbit to start generating revenue?
A: We are signing some agreements now for even the initial system we are putting up. Definitely with six satellites we can start revenue, and even with the first two we have agreements in place now with customers.
Q: Do the satellites have on-board propulsion?
A: Yes, we have a propulsion system that allows us to stay in orbit for about three years of projected life.
Q: Then atmospheric drag brings the satellites down within 25 years as recommended by debris-mitigation guidelines?
A: Exactly. We designed the system for about three years of mean mission life. After three years we will let them gracefully decline and then launch the next block. It’s really part of the system design to continually refresh. When you get the price of the satellites down low enough, and the price of the launch down low enough, you get to a point where they are disposable assets. Three years of orbital life gives you enough time to generate revenue to recoup their capital costs and replace them with an upgraded system.
It also allows you to ride atop the commercial technologies that are out there, and we use that. Even our camera is an industrial, military-grade camera and in the last year and a half there have been upgrades to that system. So we can take advantage of that in the next block. I’ve worked in space for 25 years and you are always stuck in technologies that are 10 years old when you fly them, and then up there for 15 years.
Q: Won’t launch costs be a sticking point given the short orbital life?
A: It’s true, there are some fixed costs that are pretty significant. But the launch cost has come down. Spaceflight Services has done a phenomenal job of capitalizing on that. They are pretty successful at finding available capacity.
Q: There are several companies doing something similar to what you propose. Is there a risk of oversupplying the market?
A: Many people have the impression that the field is crowded with all these start-up companies jumping in and flooding the market with systems for which there is really not that much demand. I think the opposite is true. As we talk to our customers who we’re signing agreements with now, they have been pixel-starved for years.
A lot of the algorithm and analytics firms have not been able to get off the ground because they have not been able to get enough pixels to validate all the tools and the business plans they are trying to put into place. So I think the demand is going to be well in excess of 60 satellites.
Q: Your model sounds like Planet Labs and SkyBox. What is the differentiating characteristic?
A: It really is satellite imaging as a service. We’ve seen demand from Spaceflight Services as a service, and now Spaceflight Networks as a service, with ground stations. It’s moving up the value chain.
The key differentiator between us and the others is that we intend to let our customers tap our satellites directly. They don’t get to control the satellite but they get to choose what pictures it takes.
We see a lot of existing businesses in this market are our customers. Satellite imaging companies out there could use our business to add to their existing capacity by buying images from us. We’re happy to do that. It’s just an imaging architecture that can be leased out by just about anyone.
Q: So a customer can determine the satellite’s pointing from horizon to horizon in their territory?
A: There is pretty much no limitation on the spacecraft and the orientation it can point at. We generally will restrict the imagery to plus or minus 30 degrees off-nadir, which is common in the industry. Beyond 30 degrees the imagery gets skewed and is less valuable.
Q: What is the scenario for how a customer would get access to a satellite?
A: It’s a Web-based system, the easiest way for a customer to get the imagery is to log into our website, give us a credit card and we’ve got ordering tools they can go to and determine what imagery they want, and depending on what satellite is over the area at what time, we’ll command the imagery and have it downloaded when it gets to the next ground station and then it will be distributed back to our website to the customer.
Q: What about the large business or government customer? Is there a monthly fee, or bulk pixel purchase available?
A: For a lot of these big customers it will really be machine-to-machine tasking. We will provide them with an API [application program interface] into our system so they can order directly through that API, machine-to-machine.
All of that would still be done over the Internet. But there are some customers who are talking to us about dedicated ground stations in their region for direct control of the satellite over their region. We are in discussions about providing that, but we have not set it up that way yet.
Q: So you see your customer set as everything from government and large corporations down to small businesses and even individuals?
A: Yes, even individuals can essentially get on line and have pictures delivered.
Q: What is the mass memory on board?
A: About 30 gigabytes of data storage.
Q: How many ground stations do you need?
A: We are leveraging Spaceflight Networks for ground stations and if you look at the website there are around 15, both owned and leased from others.
Q: Their network is enough for you?
A: In some cases we are putting in dedicated sites, and other places we’re leveraging existing hardware. It’s a mix of two-thirds existing hardware and one-third dedicated equipment.
Q: Is one-meter resolution sharp enough given where the market is going?
A: It’s an important question for us and we’ll be learning the answer over time. There’s a continuing debate between resolution and persistence, and which is more valuable. There is a tradeoff as a higher-resolution satellite is going to cost more and so you can afford fewer of them, and have less persistence, while the opposite is true at lower resolution.
I personally believe there is a sweet spot at about one meter in resolution. It allows you to see most human activity — cars and trucks, groups of people — but you cannot distinguish an individual person so you avoid some of the privacy issues.
So as we talk to our customers we get the sense that a 1-meter resolution satellite is a good spot, and the satellites can be inexpensive enough to provide persistence, which is going to be very valuable and something we don’t typically get from space. We see the value of persistence with UAVs. A lot of the reason you see exploding demand for UAVs has to do with persistence.