19.1 C
New York
Tuesday, May 28, 2024

See Little Robots Get Swole in This Virtual 'Gym'

If I asked you to design the perfect robot for throwing a block, you’d probably think of something humanoid, with legs for stability and hands for grasping. And who could blame you? If humans are good at anything, it’s throwing stuff.

There’s a zero percent chance you instead thought of the thing in the video above, a Frankenstein’s monster of what appear to be cobbled-together Tetris pieces. That’s because a computer “evolved” this robot’s body and the brain that controls it in a new platform from MIT’s Computer Science and Artificial Intelligence Laboratory. It’s called the Evolution Gym, where instead of relying on human designers—and their human biases—the robots of tomorrow can entrust their design to algorithms. “There's a potential to find new, unexpected robot designs, and it also has potential to get more high-performing robots overall,” says MIT computer scientist Wojciech Matusik, a cocreator of the system. “If you start from very, very basic structures, how much intelligence can you really create?”

A lot, as it turns out. Above you’ll see a soft robot that learned how to do flips. Each color-coded “voxel,” or unit, within this robot serves a purpose. Black means a rigid piece, and gray means a soft piece, neither of which do anything actively. Conversely, the colored voxels are actuators, or the bits of any robot that produce movement. (In a typical rigid robot, those are the motors that power its joints.) Blue means that the actuator contracts or expands vertically, while orange means it does so horizontally. So when you watch that robot flipping, it’s basically pivoting on the black rigid voxel at its center, while the colored actuators push it off the ground. 

This, by contrast, is a simple arch-shaped robot made only of horizontally actuating voxels. It’s legs, basically. But by contracting and expanding in coordination, the voxels make the machine move with surprising grace—kind of gallop, really.

Here’s one that’s learned how to climb. Note the blue actuators on either side of the base, which alternate their movements to get a purchase on the surface while a sort of appendage at the top of the soft robot feels its way up the column. An added challenge is that sections of the column are soft, so the machine has to adapt to these as it shimmies its way up. This is highly complex behavior for such a simple robot, much less one that designed itself. 

Most PopularBusinessThe End of Airbnb in New York

Amanda Hoover

BusinessThis Is the True Scale of New York’s Airbnb Apocalypse

Amanda Hoover

CultureStarfield Will Be the Meme Game for Decades to Come

Will Bedingfield

GearThe 15 Best Electric Bikes for Every Kind of Ride

Adrienne So

The system begins by scrambling those four kinds of voxels in unique ways to create hundreds of morphologies, in the same way that biological evolution invented the disparate forms of humans (legs), snakes (no legs), and birds (wings). In the image above, each column shows one generation of the four best-performing shapes. The algorithm also optimizes “controllers” for these robots, basically the brains that tell the actuating voxels what to do and in what order. 

Let loose to get swole in the Evolution Gym, the different morphologies—each run with different brains—are scored on how well they perform a certain task, like walking forward or throwing a block. “You pick the soft robots that perform the best, and you'll basically evolve them, or mutate them, until you refine the structures,” says Matusik. It’s survival of the fittest: The robots that walk or climb best seed the next generation, while the poor performers are tossed out. This is known as evolutionary robotics. Other researchers are using similar techniques to, for instance, evolve robot legs optimized for certain surfaces.

Here you can see the evolution of a block-carrying robot, which ends up being a galloping machine that manages to balance the object on its back. 

The designs can even adapt to obstacles in the environment, as this irregularly shaped robot shows by bounding over an uneven surface.

And here’s one with a sort of upside-down jackhammer for manipulating an overhead beam.

All of these robots have been evolved by algorithms in a simulation, so they don’t actually exist. And at the moment, these designs are limited in that they’re very focused—a particular morphology can climb or toss a block, but it can’t do both. But the Evolution Gym is an open source playground for any researcher to use and refine these techniques. (The code can be downloaded here.) 

“This is really an environment that can potentially stimulate a lot of new research,” says Matusik. “You can basically create more and more intelligent robotic systems from the ground up without any preconceived notions of what the structure or what the brain should be.” The next step after designing and training machines in the gym would be to build the best candidates in the real world. That way researchers won’t waste time piecing together a prototype only to find that it doesn’t work very well. 

​​“I am very excited to see how other researchers start using this benchmarking framework, and what type of creative virtual robots come out of it in the end,” says Tønnes Nygaard, who studies evolutionary robotics at the Norwegian Defence Research Establishment but wasn’t involved in this work. “I welcome any system that builds interest and collaboration in research, like this one, with open arms!”

More Great WIRED Stories📩 The latest on tech, science, and more: Get our newsletters!4 Dead Infants, a Convicted Mother, and a Genetic MysteryYour rooftop garden could be a solar-powered farmRobots won’t close the warehouse worker gap soonOur favorite smartwatches do much more than tell timeHacker Lexicon: What is a watering hole attack?👁️ Explore AI like never before with our new database🏃🏽‍♀️ Want the best tools to get healthy? Check out our Gear team’s picks for the best fitness trackers, running gear (including shoes and socks), and best headphones

Related Articles

Latest Articles