The M-Blocks, created by John Romanishin at MIT Computer Science and Artificial Intelligence Laboratory, are self-assembling robots that move themselves around like magnetic Mexican jumping beans. They might not look like much right now, but don't be fooled: These colorful robot-building blocks could be a technological breakthrough that makes real-life Transformers possible.
Usually, robots that can reconfigure themselves do so by sliding across one another without changing their orientation. In other words, if you had two robot cubes and marked each side with cardinal directions (north, south, east, west, up, down), those cubes could reconfigure themselves in many ways in relationship to one another, but the north sides would always point north, south sides south, and so on. This method has the benefit of allowing roboticists to always be able to model where a part of a re-assembling robot is in relation to the other parts, but it limits the ways in which robots can reconfigure themselves, and also requires a lot of complicated internal motors to get the movement right.
What makes the M-Block so unique is that it doesn't slide; it bounces. Inside every M-Block is a flywheel that spins at 20,000 rotations per minute. The momentum of this flywheel is enough to allow the M-Block to hop around, looking for other M-Blocks to attach itself to. On each side of the cube are two cylindrical magnets, mounted like rolling pins; when two M-Blocks come near each other, the magnets naturally rotate so that north magnetic poles align with south magnetic poles, or the other way around. What that means is that an M-Block can simply bounce around until it finds another M-Block, then automatically glom onto it.
This might not seem particularly useful in abstract, but it's actually incredibly exciting. Because M-Blocks can not only autonomously propel themselves but attach to one another in an infinite number of configurations using relatively cheap mechanisms, there is a lot you can do with this technology. Imagine, for example, dumping a container full of M-Blocks on an empty lot, then sending a signal for them to self-assemble into a house. Or imagine telling your couch, which is made of M-Blocks, to divide itself into three chairs instead, or transform into a bunk bed to accommodate some surprise house guests.
In fact, this is just the sort of thing that Romanishin and his team at MIT intend. They are currently hard at work building an army of 100 M-Blocks, programming them with algorithms that will help them re-assemble into increasingly complicated configurations. “We want hundreds of cubes, scattered randomly across the floor, to be able to identify each other, coalesce, and autonomously transform into a chair, or a ladder, or a desk, on demand,” Romanishin says.
Even at the current size and scale, M-Blocks have potential to be used to temporarily repair structures or provide shelter in emergencies, or raise and reconfigure scaffolding in construction. But the ultimate goal here is miniaturization: Shrinking M-Blocks down until millions of nano M-blocks can be poured out of a can and assemble themselves with nearly infinite granularity into anything you wish. Imagine 3-D printing, without the need for a 3-D printer. These bouncing magnetic robot blocks could be the technological breakthrough that gets us there.
You can read more about M-Blocks and their incredible potential here.