Episodic Learning for Safe Bipedal Locomotion with Control Barrier Functions and Projection-to-State Safety

This paper combines episodic learning and control barrier functions in the of bipedal locomotion. The safety guarantees that control barrier provide are only valid with perfect model knowledge; however, this cannot be met on hardware platforms. To address this, we utilize the of projection-to-state safety paired with a machine learning framework an attempt to learn the model uncertainty as it affects the barrier. The proposed approach is demonstrated both in simulation and on for the AMBER-3M bipedal robot in the context of the stepping-stone, which requires precise foot placement while walking dynamically.