Reproduce “FORGE: Feasible Online Reduced-Order Reference Generation and Whole-Body Execution for Agile Humanoid Locomotion” 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
Adrian B. Ghansah, Yujin An, Jeeseop Kim, Soon-Jo Chung
FORGE trains a command-responsive reference generator and whole-body tracker to connect simplified locomotion planning with humanoid execution. The first policy produces online references using kinematic end effectors and inverse kinematics; the second learns to track them with full-body dynamics. The system tracks changing commands in simulation and on a Unitree G1, reaching approximately 2.8 m/s over 15 feet; 4.0 m/s running is shown only in simulation.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.