Transformers Are Real: MIT Researchers Create Self-Assembling Robots
Have you ever, while watching Transformers, wished the shape shifting robots would appear in real life (minus the world domination struggle)? If so, a team at MIT is working on something that will be of some interest to you. Researchers at the school’s Computer Science and Artificial Intelligence Laboratory (CSAIL) are developing self-assembling and self-configuring robots. While these robots may not be as large as Optimus Prime and Megatron, they are set to accomplish great things.
M-Blocks, as the robots are called, are cube-shaped robots able to change their geometry according to task. It’s the robots’ ability to adapt to their environment that makes them so special. “This is exciting because a robot designed for a single task has a fixed architecture, and that robot will perform the single task well, but it will perform poorly on a different task in a different environment,” Daniela Rus, professor of electrical engineering and computer science and director of CSAIL, said. “If we do not know ahead of time what the robot will have to do and when it will have to do it, it is better to consider making modular robots that can attain whatever shape is needed for the manipulation, navigation or sensing needs of the task.”
Each robot is able to move itself through three main parts: first, each M-Block is equipped with a tiny magnet on each of its eight faces. This allows the robots to attach to each other in between movements. Next, each robot contains a flywheel internally, which enables the sudden bursts of spinning, jumping, and even flying motions you can see in the video above. The powerful flywheel can reach up to 20,000 revolutions per minute, which allows M-Blocks to detach and reattach themselves to the surrounding robots in any formation that is needed. Finally, the edges of each M-Block contain magnets mounted like rolling pins. When an M-Block begins to rotate, its face magnet loses contact while its edge magnet rotates to from a very strong bond with the edge of the adjacent cube. “There’s a point in time when the cube is essentially flying through the air,” said Kyle Gilpin, a postdoc working on the project. “And you are depending on the magnets to bring it into alignment when it lands. That’s something that’s totally unique to this system.”
Self-Assembling Robots at MIT
Right now, the researchers control the robots’ movement through commands sent by a radio. In the future, however, the team plans to install the algorithms on the robots themselves so the M-Blocks can choose appropriate movements autonomously.
In continued work, the MIT team plans to build an “army” of 100 cubes, each of which can move in any direction. “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 said. Future applications could include bridge or building repairs during emergencies, raising and reconfiguring scaffolding for building projects, assembling into furniture and heavy equipment as needed. One of the main benefits of the robots can work in conditions that are inaccessible to humans, and that they can work in flexible ways. “In the vast majority of other modular systems, an individual module cannot move on its own,” Gilpin said. “If you drop one of these along the way, or something goes wrong, it can rejoin the group, no problem.”
The quest to build self-assembling robots is not new. Such a simple design as that found in M-Blocks, however, is innovative. “We wanted a simpler approach that uses fewer actuators, fewer moving parts and was easier to implement on a lot of different robots,” John Romanishin, a research scientist at CSAIL and the dreamer behind the project. The few parts that make up M-Blocks will make it simpler for engineers, in the future, to design the robots in miniature.
The small and elegant M-Blocks are set to deepen humans’ ability to interact with the world around us. “It’s one of those things that you kick yourself for not thinking of,” said Hod Lipson, a robotics researcher at Cornell University and a colleague of Rus. “It’s a low-tech solution to a problem that people have been trying to solve with extraordinarily high-tech approaches.”
Images via MIT. GIF via Co.Design