Imagining Locomotion: Learning a Neural World Model for Legged Robots
A neural dynamics world model paired with model-free RL policies for quadruped and humanoid locomotion in IsaacLab, enabling long-horizon autoregressive prediction and imagined rollouts that outperform pure model-based RL in prediction accuracy, policy learning, and sim-to-real transfer.
The skinny
The skinny isn't ready yet — notes appear once the transcript is processed.