SoloParkour: Constrained Reinforcement Learning for Visual Locomotion from Privileged Experience

Elliot Chane-Sane, Joseph Amigo, Thomas Flayols, Ludovic Righetti, Nicolas Mansard

Research output: Contribution to journalConference articlepeer-review

Abstract

Parkour poses a significant challenge for legged robots, requiring navigation through complex environments with agility and precision based on limited sensory inputs. In this work, we introduce a novel method for training end-to-end visual policies, from depth pixels to robot control commands, to achieve agile and safe quadruped locomotion. We formulate robot parkour as a constrained reinforcement learning (RL) problem designed to maximize the emergence of agile skills within the robot's physical limits while ensuring safety. We first train a policy without vision using privileged information about the robot's surroundings. We then generate experience from this privileged policy to warm-start a sample efficient off-policy RL algorithm from depth images. This allows the robot to adapt behaviors from this privileged experience to visual locomotion while circumventing the high computational costs of RL directly from pixels. We demonstrate the effectiveness of our method on a real Solo-12 robot, showcasing its capability to perform a variety of parkour skills such as walking, climbing, leaping, and crawling.

Original languageEnglish (US)
Pages (from-to)4268-4285
Number of pages18
JournalProceedings of Machine Learning Research
Volume270
StatePublished - 2024
Event8th Conference on Robot Learning, CoRL 2024 - Munich, Germany
Duration: Nov 6 2024Nov 9 2024

Keywords

  • Agile Locomotion
  • Reinforcement Learning
  • Visuomotor Control

ASJC Scopus subject areas

  • Artificial Intelligence
  • Software
  • Control and Systems Engineering
  • Statistics and Probability

Fingerprint

Dive into the research topics of 'SoloParkour: Constrained Reinforcement Learning for Visual Locomotion from Privileged Experience'. Together they form a unique fingerprint.

Cite this