challenging terrain

denoising world model (1) world model learning (1) navigation (1) self-learning latent representation (1) quadruped robot (1) humanoid locomotion (1) privileged information (1) policy learning (1) sim-to-real transfer (1) reinforcement learning (1)