Humans have launched thousands of satellites into orbit, many of which are now useless and dangerously in the way of future space missions. NASA wants this space junk cleared out, but many pieces are spinning so wildly that they would be dangerous to collect. To solve this problem, a team from MIT has come up with an algorithm that could let cleanup crews measure a target's movement so they can plan an approach to safely snatch it up.
The team sent their algorithm up to the International Space Station, where astronauts tested it using two SPHERES satellites, volleyball-sized bots being tested as swarming space helpers. As one satellite floated and spun, another filmed the action using a pair of linked cameras, spaced slightly apart.
As the cameras captured the spinning satellite, which you can see in the video above, the algorithm mapped features on its surface and compared the images seen through the left and right lenses. Frame by frame, the algorithm measured slight discrepancies in distance and angle between each lens' view of the features, and used these calculations to make a 3-D map of the object rotating in space (your brain does roughly the same thing with the feeds coming from your eyes).
With enough features matched over enough frames, the algorithm can map where an object is in space, the direction it is moving, and speed and direction of its spin. The algorithm also does a pretty good job of guessing the object's center of mass, which becomes important if the object happens have a long, robotic arm.
The researchers will be presenting their work at the International Conference on Intelligent Robots and Systems, held in Chicago from September 14-18. Their full study will be published in a future issue of the Journal of Field Robotics.
NASA has around 100 satellites pegged for removal. In addition to helping to clear space junk, the authors note that the algorithm could be used during docking with a space craft that's not communicating its coordinates, or other objects, like asteroids.
