MIT Scientists Create Modular Robot Blocks That Can Self-Assemble & Reconfigure

Looking at these reconfiguring robo-cubes, created by research scientists at MIT in the face of ongoing naysaying, it strikes me that the human race can’t be far off a huge achievement: building a physical version of Tetris that self assembles. From angular chaos, to robot-enabled order. That and giving future Dalek armies the ability to bound up stairs.

The M-Blocks, shown off in the above video, are reconfigurable, modular robots with no external moving parts. The cubes’ ability to move results from harnessing the momentum of an internal flywheel (which can hit speeds of 20,000 revolutions p/m) — allowing them to climb over one another, make jumps, spin and roll around. And do all that without the need for wheels or legs.

Magnets on the corners of the blocks are used for course correction and stability, so that one small leap results in an M-Block snapping tidily into place atop its fellow, rather than going rogue and skittering uselessly off the table — although they can apparently do that, too. Chamfered edges on the cubes enhance the strength of the magnetism as the cubes rotate over each other to take up their new positions.

Reconfigurable modular robots with no external moving parts have evidently been something of a Holy Grail in the modular-robotics community. “It’s one of these things that the community has been trying to do for a long time,” says Daniela Rus, a professor of electrical engineering and computer science and director of CSAIL, speaking to MIT news. “We just needed a creative insight and somebody who was passionate enough to keep coming at it — despite being discouraged.

“Our objective is to design self-assembling and self-reconfiguring robot systems. These are modular robots with the ability of changing their geometry according to task and this is exciting because a robot designed for a single task has a fixed architecture. And that robot will perform a single task well but it will perform poorly on a different task in a different environment,” she adds in the video.

Very long term, the goal of much modular robotics research is to be able to miniaturise modules to such an extent that swarms of self-assembling microbots (or even nanobots) can be created — capable of reconfiguring themselves into different forms, shapes and sizes, and changing their function accordingly. Albeit, that’s far-off sci-fi stuff.

In the shorter term, the researchers behind M-Blocks reckon there are still potential use-cases for their more substantially sized, reconfiguring robo-cubes. They note that large numbers of the blocks could be used to temporarily repair bridges or buildings during emergencies, for instance, or raise and reconfigure scaffolding, or assemble different types of furniture or heavy equipment. Different cubes could also carry different functions — such as a camera, lights or a battery pack — to augment overall function.

The researchers are currently building an army of 100 cubes, each with the ability to move in any direction, and designing algorithms to guide them — with the aim of having the cubes transform their state from being randomly scattered across the floor, to identifying each other, coming together and then autonomously transforming into various forms (chair, ladder, etc.) on demand.