Leveraging Symmetry in RL-based Legged Locomotion Control Study

Model-free reinforcement learning can benefit from incorporating symmetry constraints to improve exploration and task performance in legged locomotion control.