Playing Soccer Like Humans Do: Learning Versatile and Agile Humanoid Soccer via Latent Motion Priors
A latent motion prior distilled from human soccer MoCap lets one skill-conditioned policy move, dribble, trap and kick with human-like motion; transfers zero-shot to a Unitree G1.从人类足球动捕数据蒸馏出潜在运动先验，使单一技能条件策略能完成移动、带球、停球和射门，动作自然流畅；零样本迁移到 Unitree G1。
BibTeX
@misc{shao2026playing,
title = {{Playing Soccer Like Humans Do: Learning Versatile and Agile Humanoid Soccer via Latent Motion Priors}},
author = {Haoyu Shao and Yunshen Wang and Jingze Zhang and Shaohang Zhu and Yong-Lu Li and Tianzhu Zhang and Baoxiong Jia and Siyuan Huang},
year = {2026},
eprint = {2611.06564},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2611.06564},
}