This is possibly one of the coolest things we’ve seen in a while. John Romanishin, a research scientist at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL), has prototyped modular robot blocks that can self-assemble.
The robots, called M-Blocks, are small cubes that, thanks to an internal flywheel and some cleverly positioned magnets on their surfaces, can move around, climb on each other, and fly through the air without having any external moving parts.
Romanishin will present a paper describing the new robots at the IEEE/RSJ International Conference on Intelligent Robots and Systems in November along with postdoc Kyle Gilpin and MIT robotics professor Daniela Rus, who collaborated with him on the project.
Fitting a brushless motor controller, a flywheel, a braking mechanism, a radio, and battery into such a small volume was a challenge, Gilpin says in an MIT video detailing the project. Also, jumping and flying through the air require high amounts of energy in a short time, something modular robots are not really designed to do.
Modular robots have practical applications in a variety of fields, including space exploration, thanks to their ability to self-configure and adapt themselves to unforeseen situations. Currently, the MIT researchers send commands to the cubes over a computer via the built-in radios. But the next step is to make the cubes more “intelligent” by directly embedding algorithms into them. Then they simply tell the M-Blocks to transform into any shape and let the cubes figure out how to arrange themselves.