Architecture and Hardware

Defending the Orbital Internet

e.Deorbit’s robotic arm.
The European Space Agency’s proposed e.Deorbit mission would use a robotic arm to catch a derelict satellite – the baseline capture method for what would be the world’s first active space debris removal mission, in 2024.

In Alfonso Cuaron's Oscar-winning movie Gravity, a U.S. National Aeronautics and Space Administration (NASA) shuttle in low Earth orbit is torn apart by a terrifying blizzard of hypersonic space debris created by an exploding Russian satellite. The blast hurls hundreds of debris fragments into multiple orbits, with shards hitting other satellites and breaking them up too, one after another, in a highly destructive, debris-generating chain reaction.

Called the Kessler Syndrome, after NASA astrophysicist Don Kessler who predicted the effect, this exponential debris collision cascade has yet to occur outside Hollywood. However, if it happens, its consequences could be catastrophic for global communications, satellite navigation, positioning, timing, and weather data. It could also render near-Earth space unnavigable by crewed spacecraft, endanger the lives of astronauts on the International Space Station (ISS), and impact future missions to the Moon or Mars.

Currently, such risks are controlled because most satellites and pieces of space debris are tracked by the radar of the U.S. Air Force Space Surveillance Network. When, for instance, a potential collision looms, the ISS can be notified to fire its thrusters to elevate its altitude to dodge any incoming debris, as it does around once per year.

So far, so good, but an emerging type of space-based Internet communications service based on "megaconstellations" comprising many thousands of satellites could  provide the trigger for a Kessler Syndrome event. That's because these megaconstellations will swell spacecraft numbers many-fold, so tracking and predicting collisions will become far tougher.

In that context, it will be vital to develop smart, autonomous satellite flight algorithms, and robotic deorbiting technologies that rid orbit of dead, uncontrollable and/or fragmenting spacecraft, to help keep the space around our world as free from debris as possible.

At issue is the fact that four companies—SpaceX of Hawthorne, CA; Boeing of Everett, WA; Samsung of Seoul, South Korea, and satcom newcomer OneWeb of Arlington, VA—have announced their intentions to launch space-based Internet services to provide online connectivity for the 45% of the world's population (some 3.4 billion people) who currently have no access to the Internet.

Today, 99% of Internet traffic travels via land and submarine cables, with just 1% traveling via expensive-to-use, low-data-capacity, high-power satellites locked in geostationary orbits (GEO) some 36,000 km. above the Earth. The aim of each of these companies is to smother the planet with arrays of non-geostationary orbit (NGSO) satellites, which are small, agile, networked communications spacecraft in far lower orbits (roughly 500 to 1,200 km) so there is always one in the sky to communicate with and, being way closer to Earth than GEO satellites, they need less power, and so require smaller receivers, too.

Greg Wyler, founder of OneWeb, a company backed by Qualcomm, Softbank, and satellite maker Airbus, aims in particular to connect the 2 million schools around the world currently without broadband to the Internet, and to provide "reliable high-speed connectivity" to poorly served rural areas of America, Europe, and Asia.

Similarly, in its U.S. Federal Communications Commission application for what it is calling its Starlink NGSO system, SpaceX says its satellite system will be "capable of delivering robust broadband service to customers around the world, especially for those in areas that find themselves the wrong side of the digital divide."

That's the good news.

The bad news: to provide NGSO services, an unfeasibly high number of satellites will be needed if each is to always be in laser-optical or radio range of each other, and in radio range of ground receivers a few hundred kilometers below. For StarLink, for instance, SpaceX plans to launch an astonishing 12,000 satellites, while Samsung plans 4,600 spacecraft, Boeing up to 3,000, and OneWeb around 600.

That all adds up to 20,200 NGSO satellites, all in the massively congested low Earth orbit (LEO) region, at altitudes between 160km (roughly 100 miles) and 2,000km (over 1,200 miles). 

If that doesn't sound terribly high, consider this: since 1957, when the space age began with the launch of Sputnik, 5,400 rockets have placed 8,650 satellites in orbit around our world. Some 4,700 of those are still in orbit (the rest having burned up), but only 1,800 of them are still working, leaving 2,900 as uncontrollable, drifting derelicts.

Because these derelict craft sometimes collide, and rocket bodies with residual fuel can explode, debris is created continually, and in the friction-free vacuum of space, it maintains its orbital velocity of at least 28,000 k.p.h. (more than 17,000 m.p.h.). There are currently 20,000 pieces of debris in orbit that are larger than a softball, and 500,000 pieces the size of a marble, according to the U.S. Air Force Space Surveillance Network.

In that context, consider what will the debris load could become when 20,200 extra satellites are added to low Earth orbit over the next five to 10 years; will it even be manageable? This is a question that has been entertaining the minds of a United Nations group called the Inter-Agency Space Debris Coordinating Committee (IADC), in which 13 national space agencies discuss debris mitigation measures.

Says Hugh Lewis, an engineer specializing in space debris modeling at the University of Southampton in the U.K., and the U.K. Space Agency's representative on the IADC, "The megaconstellation operators will likely need to perform a lot of collision avoidance maneuvers simply because of the number of satellites they will be managing. Balancing the need to maintain the coverage over the Earth with the need to maneuver to avoid collisions will be quite a challenge."

Lewis adds that while the satellite operators say the collision avoidance can be done, "it's not clear how this will be managed for constellations of thousands of satellites."

To cope with the debris problem, Lewis says operators are planning end-of-life disposal programs in which the last gasps of a satellite's fuel will be kept in reserve to push the dying spacecraft into the upper atmosphere to burn up, rather than ending up as another uncontrollable derelict in orbit just waiting to be hit by another satellite.

"The presence of dead satellites is the main driver for long-term environmental consequences. So the operators understand the need for highly reliable systems for end-of-life disposal, but it's hard to say whether their ambitions can be met on such a large scale," says Lewis.

While the operators are promising highly reliable "self disposal" of satellites as they reach the end of their useful lives, Lewis notes the operators also hope for economies of scale by using cheaper, mass-produced "cubesat" satellites to populate their constellations.

Don Kessler, after whom the syndrome is named, points out another cubesat weakness, that because the cheap satellites are small, they are also more vulnerable to being knocked out of action by very small, and possibly untrackable,  fragments of debris. 

It's not only satellites that are getting cheaper, however: more and more are going into orbit thanks to low-cost, rocket launch providers like Rocket Lab, SpaceX, and the Indian Space Research Organization (ISRO). These organizations allow massive "ridesharing" launches to be put together, flying many cubesats into orbit at once. For instance, in early December, SpaceX launched 64 satellites in a single flight of its Falcon 9 spacecraft.

To cope with the existential threat from all these satellites, European researchers in particular are testing and developing innovative ways to get dead satellites out of orbit. In September, the Surrey Space Centre in Guildford, U.K., tested the first of a number of Active Debris Removal (ADR) methods in an ISS experimental cubesat called RemoveDebris. Once released from the ISS, RemoveDebris fired a ballistic net at a drifting target and successfully ensnared it, showing that a net can be used to capture an errant piece of space junk.

The idea is that once captured, a robotic spacecraft would propel the snared target into the atmosphere to burn up. Lewis remains to be convinced, though. "The net deployed from RemoveDebris was not tethered to the main spacecraft. This would clearly need to change for a real removal attempt," he says.

In early 2019, RemoveDebris will test two more ADR methods: capturing a target with a tethered harpoon (which would pierce a non-fuel-bearing part of a satellite), and a satellite deorbiting itself by deploying a "drag sail" that uses upper atmospheric drag to slow the craft so it drops out of orbit.

These experiments, however, are mainly aimed at removing small spacecraft. What space agencies most want to get out of Earth orbit are the more dangerous potential debris producers: the very large dead satellites that, once they start to fragment as thermal effects and fuel/battery explosions tear them apart, aided by fuel sloshing due to uncontrolled tumbling, present a clear threat to swathes of spacecraft in LEO.

One such spacecraft is the European Space Agency's Envisat, an eight-ton, double-decker-bus-sized Earth observation satellite, which was launched in 2002 and quit working in 2012. Orbiting in an uncontrolled tumble at an altitude of 790km (approximately 450 miles), Envisat is perfectly placed to cause trouble for everything from the ISS to the nascent megaconstellations, should it fragment.

As Envisat will remain a risk for 150 years before it deorbits of its own accord, ESA is developing a raft of mission concepts under its "e.Deorbit" project, to work out how to capture and deorbit Envisat.

The project's first challenge, before figuring how to grab the satellite robotically, is to work out the nature of Envisat's chaotic tumble. To do this, engineers at the Instituto Superior Técnico in Lisbon, Portugal, and the University of Cambridge, U.K., have developed e.Inspector, a cubesat that would rendezvous with Envisat and send back imagery of its rotational rate and tumbling orientations.

Once the satellite's tumble is understood, ESA has worked with Airbus to develop an e.Deorbit Spacetug (basically a robot-arm-equipped satellite) which would rendezvous with Envisat, then use smart rotational motion algorithms to fire its thrusters in such a way as to match Envisat's tumble in three dimensions. Once they are stationary with respect to each other, a robot arm on the tug can securely clamp onto Envisat, and then the pair can stabilize their motion before Envisat is taken on a trip to its fiery doom.

Still another e.Deorbit plan, hatched by ESA partners GMV Aerospace & Defense of Spain, alongside Sierra Nevada Corp. of Sparks, NV, is to use the latter's Dream Chaser spaceplane, which NASA cleared for production as an ISS cargo freighter in December, to perform the Envisat ADR mission. An agile craft with multiple thrusters, Dream Chaser could match Envisat's tumble and, using robot arms that emerge from its aft end, grab and deorbit the derelict craft.

The trick will be to do all this without creating more debris. ESA hopes to firm up its e.Deorbit plans for a mission launch in 2024, around the time today's putative megaconstellations should be starting to seriously serve up Internet access from space. It's way too early to say whether such deorbiting technology will work, and if it does, who will pay for it; after all, launching a spacecraft simply to burn up another one will be an expensive option.

Says Lewis, "There is a long road to travel before we can safely contemplate the removal of satellites like Envisat; not just in terms of the technical difficulties, but also in terms of the financial ones."

These difficulties need addressing soon. In Gravity, as the Kessler Syndrome takes hold and knocks out communications satellites, astronaut Matt Kowalski (played by George Clooney) quips to his space-walking colleague Ryan Stone (Sandra Bullock) that "Half of North America just lost their Facebook."

If the orbital Internet becomes a dominant way of connecting to broadband, as many predict, a collision like that will mean we're going to lose a whole lot more than a handful of trivial social media posts.

Paul Marks is a technology journalist, writer, and editor based in London, U.K.

Join the Discussion (0)

Become a Member or Sign In to Post a Comment

The Latest from CACM

Shape the Future of Computing

ACM encourages its members to take a direct hand in shaping the future of the association. There are more ways than ever to get involved.

Get Involved

Communications of the ACM (CACM) is now a fully Open Access publication.

By opening CACM to the world, we hope to increase engagement among the broader computer science community and encourage non-members to discover the rich resources ACM has to offer.

Learn More