, Ph.D. Student, University of California Berkeley
Humanoid robots that can operate in diverse environments have the potential to address labor shortages in factories, assist the elderly at home, and colonize new planets. While classical controllers have shown impressive results in multiple settings, they're challenging to adapt to new environments. Here, we present a fully learning-based approach for humanoid locomotion. Our controller is a causal Transformer trained by autoregressive prediction of future actions from the history of observations and actions. We hypothesize that the observation-action history contains useful information about the world that a powerful Transformer model can use to adapt its behavior in-context, without updating its weights. We train our model with large-scale reinforcement learning on an ensemble of randomized environments in simulation and deploy it to the real world zero-shot. Our controller can walk on different terrains, carry loads of varying mass and shape, and is robust to external disturbances.