Self-Learning-Latent-Representation

SLR: Learning Quadruped Locomotion without Privileged Information
The recent mainstream reinforcement learning control for quadruped robots often relies on privileged information, demanding meticulous …
PDF DOI