Communication Dans Un Congrès Année : 2022

Reactive Stepping for Humanoid Robots using Reinforcement Learning: Application to Standing Push Recovery on the Exoskeleton Atalante

Résumé

State-of-the-art reinforcement learning is now able to learn versatile locomotion, balancing and push-recovery capabilities for bipedal robots in simulation. Yet, the reality gap has mostly been overlooked and the simulated results hardly transfer to real hardware. Either it is unsuccessful in practice because the physics is oversimplified and hardware limitations are ignored, or regularity is not guaranteed, and unexpected hazardous motions can occur. This paper presents a reinforcement learning framework capable of learning robust standing push recovery for bipedal robots that smoothly transfer to reality, providing only instantaneous proprioceptive observations. By combining original termination conditions and policy smoothness conditioning, we achieve stable learning, simto-real transfer and safety using a policy without memory nor explicit history. Reward engineering is then used to give insights into how to keep balance. We demonstrate its performance in reality on the lower-limb medical exoskeleton Atalante.

Fichier principal
Vignette du fichier
Duburcq 2022 reactive stepping.pdf (5.41 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-04155863 , version 1 (07-07-2023)

Licence

Identifiants

Citer

Alexis Duburcq, Fabian Schramm, Guilhem Boéris, Nicolas Bredeche, Yann Chevaleyre. Reactive Stepping for Humanoid Robots using Reinforcement Learning: Application to Standing Push Recovery on the Exoskeleton Atalante. 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Oct 2022, Kyoto, Japan. pp.9302-9309, ⟨10.1109/IROS47612.2022.9982234⟩. ⟨hal-04155863⟩
110 Consultations
461 Téléchargements

Altmetric

Partager

  • More