Reinforcement Learning

ETH Zurich Quadruped Solves Giant Rubik’s Cube With Its Feet

ETH Zurich researchers report three autonomous cube solves by a quadruped, combining RGB-D perception, planning and learned foot-manipulation skills.