c511d02265
PiperOrigin-RevId: 638447127 Change-Id: Ib1e5020a8407bc100145a6b382e985c03dd4a848
Google Barkour v0
MJCF Instructions
The MuJoCo config in assets/barkour_v0_mjx.xml was copied from https://github.com/deepmind/mujoco_menagerie/google_barkour_v0. The following edits were made to the MJCF specifically for brax:
meshdirwas changed fromassetsto..frictionlosswas removed.dampingwas changed to 0.5239.- A custom
init_qposwas added. - A sphere geom
lowerLegFootwas added to all feet. All other contacts were turned off. - The compiler option was changed to
<option timestep="0.002" iterations="4" solver="CG"/>. - Non-visual geoms were removed from the torso, to speed up rendering.
Publications
If you use this work in an academic context, please cite the following publication:
@misc{caluwaerts2023barkour,
title={Barkour: Benchmarking Animal-level Agility with Quadruped Robots},
author={Ken Caluwaerts and Atil Iscen and J. Chase Kew and Wenhao Yu and Tingnan Zhang and Daniel Freeman and Kuang-Huei Lee and Lisa Lee and Stefano Saliceti and Vincent Zhuang and Nathan Batchelor and Steven Bohez and Federico Casarini and Jose Enrique Chen and Omar Cortes and Erwin Coumans and Adil Dostmohamed and Gabriel Dulac-Arnold and Alejandro Escontrela and Erik Frey and Roland Hafner and Deepali Jain and Bauyrjan Jyenis and Yuheng Kuang and Edward Lee and Linda Luu and Ofir Nachum and Ken Oslund and Jason Powell and Diego Reyes and Francesco Romano and Feresteh Sadeghi and Ron Sloat and Baruch Tabanpour and Daniel Zheng and Michael Neunert and Raia Hadsell and Nicolas Heess and Francesco Nori and Jeff Seto and Carolina Parada and Vikas Sindhwani and Vincent Vanhoucke and Jie Tan},
year={2023},
eprint={2305.14654},
archivePrefix={arXiv},
primaryClass={cs.RO}
}