Q&A With Sir Martin Sweeting

Martin Sweeting is the founding CEO of Surrey Satellite, in England, where he pioneered the development of small, economical satellites for commercial, scientific, and military applications. He also served as cochairman of the 2008 Glasgow International Astronautical Congress. IEEE Spectrum Senior Editor William Sweet interviewed him late last year. Here are excerpts from the interview.

IEEE Spectrum: Construction of large, specialized intelligence and communications satellites has turned into a megaproblem, with proliferation of specs, excessive redundancy requirements, and cost overruns. What has tended to go wrong in the big-satellite-launching business?

MS: Because launches in space are so expensive, you fill up the launcher with a big satellite, and then because there is room to put all these different payloads on it, you do. Then the whole thing gets so costly you don’t want it to fail, so you start to get conservative. And since you now have 10 payloads rather than the 3 or 4 you started with, everything runs slower, and technology marches on. By the time you launch it, it’s like using a PC that’s 15 years old.

Spectrum: Are there lessons here for space exploration?

MS: What I’m interested in doing is looking at Earth orbit and how we can change the economics of small satellites. Can you take that model and increase the tempo of exploration?

Right now there’s an international push to explore the solar system, first on the moon and then Mars, with robotic exploration preceding human exploration. It’s encouraging that there could be 10 satellites around the moon in the next five years. But if you look at the number that are going to land, they are very few and far between. If we’re ever going to understand how to have humans there, we have to have greater frequency of exploration. And if we’re doing it in the steps we’re doing now, we’re not going to get there in two centuries. We need to speed this up, and the only way to do that is to make it lower cost.

Spectrum: What are the first steps we should take in moon exploration?

MS: Think for a moment: If we sent a couple of probes to Earth from somewhere else in the solar system and one went to the Sahara and one landed in the desert in Chile, that would be a bit unrepresentative. We need to be able to explore things more broadly. If we’re going to set up a base in the south pole region of the moon, we need to send some probes there, see if there are resources we can use, how we would prepare, how we would send larger systems to make excavations, and so on.

If we can send a series of small probes to a number of likely places, so that individually they’re not expensive—send 6 or 8 or 10 of these; if 1 or 2 fail, you still have 80 percent of the data or samples—that seems to me to be a more sensible approach. In the 32 missions Surrey Satellite has executed, we have a 94 percent success rate.

Spectrum: Can that approach be taken to Mars, too?

MS: Getting to Mars takes longer, of course, and the communications are different. But orbital stability on Mars is somewhat better than on the moon, because the gravity field is more regular. Also, you’ve got an atmosphere, which helps with controlled landing. On the moon you have to do rocket-assisted landing.

Mars is not the place we would go first. But I don’t think it’s an order of magnitude tougher. I think if we can sort things out on the moon, Mars will be twice as hard, not 10 times.

Spectrum: Could one launch clusters of small satellites to the moon or Mars, the way we now put specialized satellites in Earth orbit?

MS: Yes, that’s the plan eventually for the moon. Our last launch a couple of weeks ago [in August 2008] carried five microsatellites, each about 50 kilos, and deployed them into a constellation around the Earth. In principle, what we’re aiming for is to put an Internet around the moon to provide the communication services that all those other satellites and landers are going to need.

Spectrum: Could it turn out that a large number of robotic missions on Mars could be more cost-efficient than a single very expensive and risky sample-return mission?

MS: Yes. For example, you could take advantage of Mars’s atmosphere. You could place an airship in it and trundle around, either landing and sampling or dropping off little probes. Why not have a balloon or some sort of aero vehicle? You could have a small little insect thing that flies around or hops or buzzes and goes from place to place on the Martian surface. Or it could be a larger vehicle like a zeppelin that drifts around in the wind, dropping off little robotic ants that go and look but don’t come back—they just take samples and send back data. That might be better than firing lots of darts into the surface of Mars or landing on it.

On the moon you can’t do that. It’s more likely on the moon we have to have little landers. Mars just might be the place where you have a mother ship dropping off little satellites.

Spectrum: Do you see growing support in the United Kingdom for developing the autonomous ability to put humans in space?

Martin Sweeting: The UK’s position, at present, is that we’re not going to take a lead in manned spaceflight. We do want to support it, but by providing infrastructure and by playing a role in robotic exploration. The UK has a strong record in communications, navigation, and robotics. So we’ll play to those strengths.

Spectrum: Given the known pitfalls of huge, ambitious space projects, does it make sense to be setting the human exploration of Mars as our ultimate objective?

MS: I think it does make sense, but I also think we have a responsibility to put that ambition in context. Look at what happened with the moon: We went there and then just gave up. Where’s it leading? Why are we doing it? What are the possibilities?