Spysats for Everyone
Swarms of small, orbiting cameras are coming. To watch.
- By Tony Reichhardt
- Air & Space magazine, September 2013
(Page 2 of 3)
Now anyone will. Starting next year, you’ll be able to watch online, for free, HD videos from UrtheCast’s cameras almost as soon as they’re sent down from the ISS. The company will pin the videos to a base map that will fill in over time with all kinds of scenes, from momentous to trivial. “So you might see the president’s inauguration, the Springsteen concert, the Super Bowl,” says Larson. “Some things will have no social and redeeming value, just sort of silly things. Others will be more important—say, the riots in Egypt’s Tahrir Square or Hurricane Sandy. Our hope is that these videos go completely viral, that they create a unique kind of water-cooler event for people around the world.”
Having already raised $55 million, UrtheCast recently joined the Toronto Stock Exchange, and Larson expects his staff of 30 to grow to 80, with new offices in Russia and the United States. They already have an office in San Francisco, where the all-important Web interface is being designed.
Compared to past remote sensing ventures, the new ones have hardware costs that are absurdly low. UrtheCast’s first two cameras, built by the Rutherford Appleton Laboratory in England, cost $15 million. Digital Globe’s GeoEye-1 satellite, launched in 2008, cost nearly half a billion dollars to build and launch, and GeoEye-2, which was planned for launch this year but is now on hold, was even more expensive.
Larson doesn’t claim UrtheCast’s videos will compete with Digital Globe’s high-resolution pictures. Still, he says, “we will collect obscene amounts of data, obscene amounts. Our vision is to cherry-pick some of it for customers that want to buy it. And enormous amounts of it we’ll put up on our website for free.” He’s not sure exactly how people will use it, or what secondary applications might develop when they start playing with the raw data. “Candidly, we don’t know where this is going to go. We’re going to experiment."
Another player in the new market, Skybox Imaging of Mountain View, California, plans a constellation of 24 small Earth-imaging satellites, each weighing just 220 pounds, about 1/20 of what GeoEye-1 weighs. According to the company’s website, the satellites will “capture high-resolution imagery and the first-ever HD video of any spot on Earth, multiple times per day.” Skybox plans to start launching on a Russian Dnepr rocket by the end of the year.
Maybe the most radical of the new wave of remote sensing ventures is Planet Labs, based in San Francisco. By the end of the year, the company intends to launch 28 of its Dove satellites, each the size of a toaster oven, into a low orbit similar to the space station’s, where they’ll be strung out, evenly spaced, like pearls. They’ll overlook most of the populated world below 52 degrees latitude (from London to the tip of South America), with frequent revisit times for any one location as Earth turns below (the company is still working out the details, and won’t disclose the exact frequency).
The 12-inch cubesats are packed with cheap but capable technology, like high-quality solar cells scavenged, literally, from scraps on the factory floor where larger, more expensive arrays are made, and X-band radio packages adapted from Wi-Fi components. The cameras use off-the-shelf optics, which, on such a tiny spacecraft, can manage only images with a resolution of three to five meters. That’s okay with Planet Labs. By limiting their first generation of satellites to medium resolution, they sidestep privacy concerns. “Users are able to see the canopy of a tree but cannot identify individual cars or persons,” the company is careful to point out on its website.
Steve Jurvetson of the Silicon Valley venture capital firm Draper Fisher Jurvetson, who has made Forbes magazine’s “Midas List” of tech investors three years in a row (a major backer of Elon Musk’s SpaceX and Tesla, he owns the first Tesla Model S to roll off the assembly line), joined other investors in a $13 million round of financing for Planet Labs. Jurvetson says his firm looked at many of the new Earth imaging startups and was attracted to Planet Labs for several reasons. One was the team. Two of the founders, Will Marshall and Chris Boshuizen, came from NASA’s Ames Research Center in Silicon Valley. There they worked on small satellite projects, culminating in the launch of three “PhoneSats”—tiny spacecraft built around barely modified Android phones, launched last April to test how far you could push the DIY, off-the-shelf approach to space hardware. Pretty far, it turns out. The PhoneSats—named Alexander, Graham, and Bell—worked fine, and even took decent pictures from orbit.
Marshall and Boshuizen, along with partner and fellow NASA alum Robbie Schingler, brought that same mindset to Planet Labs, which they formed (originally as Cosmogia, Inc.) in 2010. The first two prototype Doves were launched in April, within two days of each other, on two different rockets. Both performed flawlessly. Marshall says that when he showed the test pictures to photo-reconnaissance experts, including some with spysat experience, they were “absolutely astonished” at the quality produced by such a tiny satellite. “They saw it previously as almost a toy,” he says.
Jurvetson is also attracted to Planet Labs’ original approach to designing constellations of satellites. With cheap, assembly-line spacecraft and launch costs for the smallest ones around $300,000 and dropping, you can start to think in terms of what he calls “disposable satellites.” So what if the Dove “flock”—yes, that’s what it’s called—orbits so low that atmospheric drag will bring the satellites down sooner than typical Earth-viewing spacecraft (GeoEye-1 is nearly 200 miles higher)? Low orbit is a good thing. It gives you better resolution with a small camera. It solves the orbital debris worry—no need to actively deorbit the Doves so they don’t add to the space junk problem. And because the satellites have to be replaced regularly, the technology stays current.