Boston, MA. About 5,000 satellites have launched since the dawn of the space age, and about 1,000 are still operable. Almost all of them were not built to be serviced in orbit. (The Hubble is an exception.) Nevertheless, NASA wants to find a way to bring orbit-side service to satellites that may run out of fuel.
Speaking Thursday at the Embedded Systems Conference, Benjamin Reed, deputy program manager, Satellite Servicing Capabilities Office, NASA’s Goddard Space Flight Center, said the goal is to make the most of your tax dollars. “I work for you,” he said. “NASA’s job is to serve the nation.”
The median satellite in operation today, he said, is about the size and weight of a minivan, costs $250 million, and takes four years to build. Each is technologically obsolete the day it launches, he added, yet it will be used for 15 years. He sees in-orbit service capability as a force multiplier to make NASA more efficient in pursuit of its two missions: science (looking for life on other planets, for example) and exploration (getting boots on other planets). Service, he said, is a way to make the most of a satellite’s initial $100 million ride into space.
With repair being a difficult challenge, redundancy offers an alternative. But the logic doesn’t work, Reed said. How many copies do you need—three, four? “You just can’t launch enough stuff,” he said—hence the logic of repair and replace with the capability to upgrade, assemble, replenish consumables, and build in space.
NASA’s current initiative at in-orbit service is the Restore-L technology demonstration mission for Landsat. Landsat 7 and 8 are now in operation, and Landsat 9 is being built to replace Landsat 7. Unfortunately, Landsat 7’s fuel may not last until 9 is launched, and even if it does, it’s worth keeping 7 operating to complement Landsat 9 by collecting even more data.
Cars are built for refueling, and no training is needed to refuel one. Not so with satellites. Service involves difficult operations: rendezvous, grasp, refuel, and relocate. These are six degree-of-freedom operations that must take place with no human in the loop. (NASA will actually use 7-DoF robotics, Reed said.) Particularly troublesome is finding something sturdy to grasp, as the satellite is covered with a gold blanket that’s taped on. “I hate tape,” Reed said. Velcro is a much better choice.
In addition, cars have gauges to tell you how much fuel is left. Not so satellites—they don’t measure fluid flow, but make estimates based on the number of thruster firings. As the service paradigm develops further, it will be necessary to measure fuel use.
The Restore-L mission involves simulations on the ground with industrial robots. Next up is Raven, which will perch on an outside platform of the International Space Station to demonstrate elements of the real-time, relative navigation technology that would be needed for a refueling system. Raven is part of Space Test Program-Houston 5, a complement of 13 unique experiments from seven government agencies that is integrated and flown under the management of the Department of Defense’s Space Test Program.
Reed touched on some of the embedded-system aspects of the mission. It relies on SpaceCube, a cross-cutting, in-flight reconfigurable on-board hybrid science data processing system based on Xilinx radiation-tolerant Virtex 5 FPGAs. The goal of SpaceCube is to provide 10x to 100x improvements in on-board computing power while lowering relative power consumption and cost. The SpaceCube design strategy incorporates couples the FPGA technology with an integrated upset detection and correction architecture to provide reliable order of magnitude improvements in computing power over traditional fully radiation-hardened flight systems.
NASA plans to launch Restore-L in late 2019.
In a question-and-answer session after the presentation, Reed said NASA is working to make future satellites more serviceable, with high-contrast black-and-white decals that can serve as targets for service satellites, and tape is being replaced with Velcro.
When asked if it would possible to “service the servicers,” he said it may be possible to refuel them, particularly for servicers in geostationary orbit. While Restore-L is a low-earth-orbit mission, it may ultimately be possible to launch a service satellite and separate fuel supply into geostationary orbit, with the service satellite returning to the fuel supply between visits to multiple geostationary satellites.
