Learning to Walk from Three Minutes of Data with Semi-structured Dynamics Models

We present a novel framework for learning predictive models for contact-rich systems which seamlessly integrates structured first-principles modeling techniques with black-box autoregressive models. This semi-structured approach enables us to make accurate predictions far into the future with substantially fewer training samples than prior methods. We leverage this capability to push the sample-complexity boundary for real-world model-based reinforcement learning. We validate our approach through real-world experiments with a Unitree Go1 quadruped robot, learning dynamic gaits -- from scratch -- on both hard and soft surfaces with just minutes of data.