Abstract: Although humanoid and quadruped robots provide a wide range of capabilities, current control methods, such as Deep Reinforcement Learning, focus mainly on single skills. This approach is ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results