Facebook's AI Teaches Robots to Adapt Rapidly to Real-World Scenarios

A team of researchers from Facebook AI, UC Berkeley, and Carnegie Mellon University’s School of Computer Science revealed Rapid Motor Adaptation (RMA), their artificial intelligence project that makes legged robots adapt intelligently in real-time to unfamiliar terrains and circumstances. The new system is based on two key components: a base policy trained through reinforcement learning (RL) and an adaptation module trained using supervised learning. The team noted that one of the most important advancements this time is the robots' ability to adapt to factors in their environment: AI allows adapting to the weight of a backpack, the amount of friction on a new surface, and more.

RMA is described as "the first entirely learning-based system to enable a legged robot to adapt to its environment from scratch by exploring and interacting with the world." Facebook states the new AI "outperforms alternative systems when walking over different surfaces, slopes, and obstacles, and when given different payloads to carry."