Offline Adaptation of Quadruped Locomotion using Diffusion Models
This addresses the challenge of offline adaptation for quadruped locomotion, enabling more flexible and efficient robot control in real-world applications.
The paper tackles the problem of enabling quadruped robots to learn and adapt to multiple locomotion skills offline, presenting a diffusion-based approach that extracts goal-conditioned behaviors from unlabeled data and runs efficiently on onboard hardware. They demonstrate the framework's validity through hardware experiments on the ANYmal platform, achieving compatibility with multi-skill policies with minimal compute overhead.
We present a diffusion-based approach to quadrupedal locomotion that simultaneously addresses the limitations of learning and interpolating between multiple skills and of (modes) offline adapting to new locomotion behaviours after training. This is the first framework to apply classifier-free guided diffusion to quadruped locomotion and demonstrate its efficacy by extracting goal-conditioned behaviour from an originally unlabelled dataset. We show that these capabilities are compatible with a multi-skill policy and can be applied with little modification and minimal compute overhead, i.e., running entirely on the robots onboard CPU. We verify the validity of our approach with hardware experiments on the ANYmal quadruped platform.