Today's satellites are all but unchangable once they're in orbit. Here, the Defense Meteorological Satellite Program.

Today's satellites are all but unchangable once they're in orbit. Here, the Defense Meteorological Satellite Program. Air Force

Here's What the Pentagon Wants Its Satellites to Do in 15 Years

The ability to take on new missions will help tomorrow’s constellations survive war in space.

As Chinese anti-satellite weapons reach ever higher, the U.S. military is getting serious about satellite constellations that can absorb combat damage and keep transmitting data.

It’s been almost a decade since China startled the world by destroying a weather satellite in low-Earth orbit about 500 miles up; two years ago, a Chinese test rocket reached an altitude of 6,250 miles. Increasingly, the barriers to attacking the most valuable satellites — the ones 22,300 miles up in geosynchronous orbits that keep them over a single point on Earth — are geopolitical, not technological .

Potential responses to the threat include cheaper satellites that you can throw up quickly as replacements, or mobile satellites that try to dodge when someone shoots at them. But Pentagon officials are increasingly looking at satellites that can fill in the gaps when a sister satellite gets taken out.

That means that future communications satellites need to be ready to do a lot more than they do today, says Winston Beauchamp, the Air Force’s deputy undersecretary for space. They need to operate on a broader range of frequencies than today’s mostly high-frequency military communications satellites. They should use software-defined radio to generate multiple waveforms. They should carry configurable antennas with variable power. And they should be able to work suddenly with other sats to do their sensing (synthetic arrays) or to form new networks for collecting and broadcasting signals.

In other words, they should be able to perform a much wider variety of operations with almost no planning beforehand. “You have a modular capability with multiple options on board and which … either [singularly] or in combination could be put to use in many, many different ways,” Beauchamp said in March at a satellite conference.

Modular capability really just means more functions. Think of how a modern smartphone allows its owner to make calls, get directions, check email, play Candy Crush, and log calories. The problem, says Beauchamp, is that our current satellites “are optimized for one mission or another,” the equivalent of boring old phones with no special features.

Building modular satellites is easier said than done. One of the bigger challenges is thermal control, said Gordon Roesler , a program manager at the Defense Advanced Research Projects Agency, or DARPA. “You have no air in space and you have a box of electronics. It will inevitably generate some heat and, now, how do I get rid of that? There are a number of ways to do that but they all get harder if you start trying to make this out of Legos.”

Another obstacle is figuring out how to connect power and data wires between the modules. Called the harness, this tangle of connections is complicated enough in a single-purpose satellite. “What they do today is design the whole satellite [and] say, ‘This box is going to go here. This box is going to go here.’ Then they figure out where all the wires go. The wire harness for a typical communications satellite can weigh over 100 pounds. If you try and make it modular it might grow by a couple hundred pounds just because the wires have to be redundant. You don’t exactly know what’s going to hook where at any given time. So those two problems for truly modular satellites are unsolved,” Roesler said.

Arming Satellites, Literally

If it’s too hard to anticipate everything a satellite might do before it goes up, you can build it to accept modifications and upgrades once in orbit. To that end, DARPA has an active program called Robotic Servicing of Geosynchronous Satellites, or RSGS . The concept is to put a robot arm on a regular commercial satellite, which would then make “house calls in space” to other satellites in geosynchronous orbit.

“Two satellites will move close together and then, when they’re about two meters apart, the robot arms that we’ve developed will reach out and grab the ring that was used to attach the satellite to its booster. That ring is exposed and very strong. So that’s what will dock. Now that we’re docked, we’re not moving, so it’s easy to go fiddle around with solar panels, do very close inspections without cameras,” said Roesler, who manages the program at DARPA.

In that docked position, the service satellite will inspect the other satellite, push it into a new position, or install upgrades to reflect innovations on the ground. “Even fully functional satellites sometimes find their working lives cut short simply because they carry obsolete payloads—a frustrating situation for owners of assets worth hundreds of millions of dollars. With no prospects for assistance once in orbit, satellites destined for GEO today are loaded with backup systems and as much fuel as can be accommodated, adding to their complexity, weight and cost,” writes DARPA in a release.

Here’s a video that explains the program:

Space-based servicing might even help satellites deploy delicated phased-array antennae that can handle a broad range of wavelengths. It’s hard to design a phased-array antennae that can survive a launch, but perhaps one could be built in orbit by a service satellite working from a bag of parts. “That’s a capability we think we can bring. It’s not one of our initial design criteria, but we think we can do it,” Roesler said.

But in order for that robot to do its work, commercial satellite makers need to design in some sort of maintenance or upgrade port or hatch.

“Today, I talk about this mechanical attachment that provides you with a limited suite of things you could do, but if you were able to connect it to the power and data bus of the customer satellite when you put this module in, that box, that 100-kilogram box, it’s going to have even more capability,” said Roesler.

“Once satellite manufacturers start putting the equivalent of a USB port on their satellites, so my robot could bring the equivalent of a new thumb drive… that would enable even more capabilities” he said. “I’ve spoken to many manufacturers and they’re considering it. [The unarmed satellite] wouldn’t have to bring its own power, for example. People are thinking about putting this on their spacecraft so it would be easy to refuel.”

Put all of those pieces together and a picture of the military satellite of the year 2030 emerges: it’s modular and designed to be upgraded or modified in space. The biggest change, in Roesler’s view, is that it isn’t actually a military satellite at all. The commercial sector, with more than enough bandwidth available to meet all of the military’s space and based communication needs, will fulfill that role.

Satellites, after all, are just another piece of information technology, and so they’ll follow the same innovation trajectory of all that came before them. Consider the integrated circuit, a technology that the military, through Texas Instruments , developed for the Minuteman ICBM. Today, the commercial sector controls the integrated circuit market and the military is happy to buy them commercially at low cost and high volume. “The same thing will happen with telecommunications satellites,” said Roesler. “We have some influence now. We build some of our satellites. Ten or 15 years from now, we would never imagine doing that.”