Geosynchronous orbits above Earth are among the most valuable real estate in the solar system. This band of space is utilized by everything from civilian communications and GPS satellites to government-operated weather and nuclear monitors to military applications like on-demand warfighter broadband. It's also a veritable minefield of broken-down, ground-up derelict satellites.
As Dr. Darren S. McKnight of Integrity Applications explained during a recent presentation at the 32nd Space Symposium held in Colorado Springs, Colo., this week, every satellite collision could potentially produce hundreds to thousands of debris fragments. And each of those fragments in turn becomes a potential satellite-killing missile. Even tiny bits of debris just a centimeter in diameter, known as the lethal non-trackable (LNT) population, can blast holes clean through satellite components, rendering the spacecraft non-operational.
In fact, these LNT debris are in many ways more dangerous than larger pieces, due to the sheer number of them. McKnight calculates that there are anywhere from 15 to 30 times as many LNT debris currently in orbit than the entire cataloged population of pieces bigger than 10 centimeters.
"In an absolute sense, there is a 10 to 50 percent chance that at least one of each of the constellation members will be struck by a one centimeter fragment (or larger)," McKnight wrote in a paper published last year. "This would likely terminate that satellite's mission and might even liberate debris that would pose additional hazards." The probability of getting hit by a piece of debris 10 centimeters or larger, however, is just under 3 percent.
I spoke with Naval Research Laboratory aerospace engineer Bernard Kelm at the Space Symposium. Turns out, repairing satellites in orbit is just as much of an engineering challenge as it sounds. "Every satellite is different. They're all custom devices built for their own custom purposes," Kelm said. However, despite their custom-built nature, there really isn't that much variation in terms of components when it comes to satellites -- thanks in large part to the limited number of manufacturers in the industry.
Take the satellite's launch interface, for example. It's the bit on the satellite's backside that connects it to the rocket during launch and almost exclusively uses a "cup-cone" setup. That's where a small knob on the rocket mates with an indentation ring on the back of the satellite to hold it in place. "There's only two ways it can be attached," Kelm said. "It's a ring clamp with only a few profiles; we can have one tool that services them all. And that's what we've done since 2002: We've partnered with DARPA to design one servicer that can service multiple vehicles."
An even bigger challenge will be getting the RSV to do what it needs to do, when it needs to do it, on the first attempt if at all possible. First, there's about a three-second delay between the ground station communications and satellites in geosynchronous orbit. Under normal conditions, this wouldn't be much of an issue. However, when you're trying to grab ahold of another satellite while both craft are tumbling around 22,000 miles above the planet, three seconds might as well be three days.
So, to account for the signal delay, the team has crammed as much processing power as it can into the service satellite itself. Or at least as much as the limitations of current-gen spaceflight computers allow. "They have orders of magnitudes less power than what you have on your desktop," Kelm said. In fact, the first RSV iteration will be among the most processing-intensive vehicles to fly.
What's more, the added autonomy will enable the RSGS satellites to operate more safely. The NRL team is pre-programming a wide array of potential issues, failures and SNAFUs into the service satellite's database. That way, if something goes awry during a docking attempt, we're not just handing an engineer the (three-second-delayed) joystick to a multimillion spacecraft that's zipping around the edge of space. Instead, the RSV will automatically halt the maneuver (potentially before ground control is even aware of the issue), drop into a predefined emergency routine and await for operators to make a final abort/continue decision.
A second issue facing the RSGS program is that robotic arms generally lack the dexterity of human appendages, despite often having the same degrees of freedom. "A robot arm is a very accurate but very rigid platform," Kelm said. This creates issues the moment the service satellite's two-meter grappling "FREND" arm comes in contact with the client satellite in zero gravity because the imparted force will cause the two to drift from each other. Granted, they won't go rocketing away thanks to the dampening effects of inertia, but it's still enough to cause havoc if ground control can only see the situation from three seconds in the past. As such, the NRL team added a force torque sensor in the wrist so that the arm can meter its approach and grab the client spacecraft without knocking it out of orbit.
These technological advancements will enable the RSV to perform four primary mission types: ultra-close inspection, anomaly resolution, orbit modification and, eventually, external upgrades. Close inspection jobs will be the first and most straightforward that the RSGS program takes on. Putting the service satellite's camera mere inches away from a troubled client satellite will give ground control an unprecedented view of the potential issue. Anomaly resolution will takes inspections a step further by allowing the RSV to physically manipulate other spacecraft. That way if a satellite's solar panel, say, gets caught a wire while unfurling, the RSV can simply poke the obstruction out of the way rather than have the entire mission be scrapped.
Orbit modification is a bit trickier. When a client satellite is unable to make the jump to a different orbit -- or even maintain the orbit its currently in -- due to insufficient fuel, the servicer would sidle up to the client spacecraft, inject propellent and send the satellite on its way. "That's a relatively expensive mission for us," Kelm said. "It'd use a lot more fuel than other missions but we will have that capability."
The fourth mission type is easily the NRL's most ambitious. The agency wants to use RSVs to install standalone modules with new capabilities on existing satellites. So, say the NOAA invents a brand-new ocean-level sensory system or the Department of Defense creates a new laser-based communications array. Instead of spending years and billions of dollars developing brand-new satellites to haul these modules into orbit, the NRL wants to use an RSV to attach it to a satellite that's already in space. It is not going to replace the functionality of a large satellite but is an inexpensive and relatively practical means of installing new functions on existing satellites. "We hope that once we prove the capabilities of this method, that satellites will begin being designed and built with docking ports for taking upgrades," Kelm said.
Most impressive, perhaps, is the fact that the NRL doesn't envision constellations of these service satellites going up in 20 years or even 10. The lab figures that with a suitable commercial partner, it can have the RSGS program in operation by the end of the decade -- five years, tops. "[Commercial industry] has 10 times as many satellites, and 10 times the need for these services," Kelm concluded. "There will come a time in the future that there will be a national need for this."
The NRL clearly isn't the only agency that thinks so. Vivisat -- a joint firm created by U.S. Space and Orbital ATK -- announced its Mission Extender Vehicle back in March. This spacecraft will perform the same sort of third mission orbit modification work as the RSGS program. And at the Symposium this week, the company revealed that its first commercial customer will be Intelsat. Orbital ATK expects to have the MEV refueling satellites in orbit by 2019.