Researchers at the Massachusetts Institute of Technology (MIT) will soon present a paper describing algorithms that could eventually enable “smart sand”; self-sculpting sand that can take on any shape. Their research involves cubes measuring approximately 10 millimeters (0.4 inch) with simple microprocessors embedded within, and magnets affixed to four of its sides. It is believed that the sand will be able to spontaneously form new tools, or even duplicate broken mechanical parts.

smart sand, smart robotic sand, MIT, Massachusetts Institute of Technology, Daniela Rus, Kyle Gilpin, reconfigurable robots, self sculpting sand, 3Dprinting

Envision that you have a huge box of sand. You dunk in a dollhouse-sized chair only to reach down into the box to pull out a full-sized chair for your office or dining room. This is the idea behind the project at MIT’s Distributed Robotics Laboratory. Thanks to complex algorithms, the tiny grains of programmed sand could be informed which way to bond and what shape to take as you will.

The creation of an object or spare part out of smart robotic sand would be a subtractive process similar to stone carving or sculpting — unlike other reconfigurable robots that use an additive method like clicking LEGO blocks together. Each individual grain of sand would transmit messages back and forth and selectively bond to create a three dimensional object. The grains not utilized would simply fall back into the sand pile.

MIT computer science professor Daniela Rus and student Kyle Gilpin are the brains behind the study, and the duo is starting with engineered cubes, or “smart pebbles,” to test their theory. Each pebble has 32 kilobytes of code and two kilobytes of programable memory, and the pebbles use “electropermanent” magnets that can be pulsed on and off quickly to maintain their pull on each other — they do not require a constant electrical current.

While it will be a while before the self-sculpting sand can be seamlessly shaped into a kitchen stool or a spare part for a computer printer, Rus and Gilpin do have confidence in their work and their peers are taking notice. The pair will present their paper at the IEEE International Conference on Robotics and Automation, the world’s leading robotics conference, in May.

Via MIT News Office

Photos from MIT courtesy M. Scott Brauer of MIT Media Relations, beach sand Leon Kaye