Reproduce “Playing Soccer Like Humans Do: Learning Versatile and Agile Humanoid Soccer via Latent Motion Priors” 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
Haoyu Shao, Yunshen Wang, Jingze Zhang, Shaohang Zhu, Yong-Lu Li, Tianzhu Zhang, Baoxiong Jia, Siyuan Huang
SOCCERBOT uses one skill-conditioned policy to control a humanoid across moving, trapping, dribbling, and kicking. It adapts a shared repertoire distilled from soccer motions to each command, with a prior anchoring movement and a task policy adding adjustments. In simulation, it achieves 97.04% kicking and 89.88% dribbling success, but trapping success is 83.08%, and hardware transfer varies by skill.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.