Reproduce “GenLoco: Generalized Locomotion Controllers for Quadrupedal Robots” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Gilbert Feng, Hongbo Zhang, Zhongyu Li, Xue Bin Peng, Bhuvan Basireddy, Linzhu Yue, Zhitao Song, Lizhi Yang and 3 more
GenLoco trains quadruped locomotion policies across procedurally varied morphologies, using sensor and action histories without explicit morphology input. On Mini Cheetah pacing, it achieved 0.743 ± 0.092 normalized return over six trials, while the robot-specific policy failed all six. This supports transfer within tested similar quadrupeds, not arbitrary robots.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.