Carnegie Mellon University
July 09, 2021

AI Allows Legged Robots To Adapt in Real-Time to Changing Conditions

By Aaron Aupperlee

Aaron Aupperlee
  • School of Computer Science
  • 412-268-9068

Researchers from Carnegie Mellon University, the University of California, Berkeley, and Facebook AI didn't just teach a robot to walk — they taught it how to learn to walk.

The distinction is key. A major hurdle to deploying legged robots, whether with two, four or even more legs, is figuring out how the robot will respond to changing conditions. Humans can adapt as they walk over rocks, mud, sand, slippery ice and uneven surfaces. They adjust to carrying a heavy backpack or limp along with an injured ankle.

Legged robots cannot adjust so quickly. Most legged robots must be hand-coded for their environments. A crack in a sidewalk or a patch of oil can stop a robot in its tracks or cause it to come tumbling down.

Rapid Motor Adaptation (RMA) seeks to change that. The artificial intelligence was jointly developed by Deepak Pathak and Zipeng Fu at CMU's School of Computer Science and Ashish Kumar and Jitendra Malik at Berkeley AI Research. It enables legged robots to adapt intelligently in real time to challenging, unfamiliar new terrain and circumstances.

"The focus is not walking. It is learning," said Pathak, an assistant professor in the Robotics Institute at CMU. "By falling thousands of times or millions of times in simulation, it learns to walk from scratch and adapts to the ever-changing real world.

"Since the algorithm's focus is learning, it is applicable to any kind of robot, not just this one."

The Rapid Motor Adaptation algorithm helps four-legged robots adjust in real-time to unseen scenarios such as changing terrains, changing payloads, wear and tear.

RMA is the first entirely learning-based system that does not rely on any hand-coded motions. and allows legged robots to adapt to their environment by exploring and interacting with the world. Testing showed that robots with RMA outperformed competing systems when walking over varied surfaces, slopes and obstacles, and when carrying different payloads.

"If you pick up a backpack, you adjust your motion without knowing the exact weight. If the terrain beneath your feet changes, you adjust your balance to compensate. RMA does this by adapting the robot joints in real-time," said Kumar, a Ph.D. student at Berkeley.

The technology isn't limited to robotics. RMA is a step toward building AI systems that can learn in real time to adapt to changing and challenging conditions. The team will present their research, "RMA: Rapid Motor Adaptation for Legged Robots" at Robotics: Science and Systems, a conference held virtually this month. More information about RMA and its development is available in this blog post from Facebook AI.

— Related Content —