Text this: A reinforcement learning approach to quadruped locomotion: framework, training, and evaluation