Research labs
Improbable AI Lab
United States
Rapid Locomotion via Reinforcement Learning reached record speeds of 3.9 m/s on the MIT Mini Cheetah and transferred to the Unitree Go1 across natural terrain.
AffiliationMIT CSAIL
CountryUS
TypeAcademic
Research areasdexterous manipulation, agile locomotion, reinforcement learning, sim2real
Flagship contributionDemonstrated Visual Dexterity, real-time in-hand reorientation of novel and complex object shapes from a single depth camera on a hardware setup costing under 5,000 dollars.
Key peoplePulkit Agrawal (Steven and Renee Finn Chair Assistant Professor, MIT EECS, directeur du lab), Gabe Margolis (doctorant, auteur principal de Rapid Locomotion)
Open outputsCode de Rapid Locomotion via Reinforcement Learning (GitHub Improbable-AI/rapid-locomotion-rl) ; publications arXiv en accès libre pour Visual Dexterity.
CityCambridge (Massachusetts)
Sources : Improbable AI Lab (2026-07)arXiv (Science Robotics 2023) (2022-11)arXiv (RSS 2022) (2022-05)
← Back to comparator · Research labs