Learning vision-driven reactive soccer skills for humanoid robots
Science Robotics · 19.08.2026 · 11 (117)
DOI: 10.1126/scirobotics.aed1152 ↗Kurz gefasst
Fachartikel in der Zeitschrift Science Robotics, erschienen am 19.08.2026. Verfasst von Yushi Wang und 9 weiteren Autor:innen. Beteiligte Einrichtung: Department of Automation, Tsinghua University, Beijing 100084, China. Hauptthemen laut MEDLINE: Soccer, Robotics, Visual Perception. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 65 Literaturangaben. Gefördert von: Beijing Natural Science Foundation, Science and Technology Innovation 2030 Major Project, Tsinghua University Initiative Scientific Research Program.
Abstract (Originalsprache)
Humanoid soccer poses a representative challenge for embodied intelligence, requiring robots to coordinate agile locomotion with unreliable visual perception in dynamic environments. However, existing systems typically rely on modular pipelines that separate perception from control or assume ideal sensing, making it difficult to achieve coherent and reactive behavior under real-world perceptual limitations. In this work, we present a unified reinforcement learning–based controller that enables humanoid robots to learn vision-driven reactive soccer skills by directly coupling visual perception with locomotion control. The robot is trained in simulation to acquire soccer behaviors, and adversarial motion priors guide policy learning toward natural motion patterns. To support robust performance under imperfect sensing, we introduce an encoder-decoder architecture together with a virtual perception system that models key characteristics of onboard vision, exposing the policy to perceptual noise and detection failures during training. This design encourages the policy to internalize perceptual uncertainty and continuously adapt its motion in a closed loop. The resulting controller produces coordinated soccer behaviors using only onboard vision, including ball searching, chasing, and multidirectional kicking. It reduces ball position estimation error by 46% and shortens time-to-kick by up to 64% compared with a rule-based baseline, achieving around 90% kicking success in frontfield positions. Experiments across diverse environments and dynamic scenarios, including real RoboCup competitions, further demonstrate the robust performance of the controller. These results highlight the practical effectiveness of integrating perceptual uncertainty directly into policy learning for achieving reliable vision-driven behaviors in humanoid robots operating under real-world conditions.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Department of Automation, Tsinghua University, Beijing 100084, China.
- ByteDance Seed, Beijing 100080, China.
- College of Engineering, China Agricultural University, Beijing 100083, China.
- Beijing Key Laboratory of Embodied Intelligence Systems, Beijing 100084, China.
- Institute for Embodied Intelligence and Robotics, Tsinghua University, Beijing 100084, China.
ORCID-Profile: Yushi Wang, Changsheng Luo, Penghui Chen, Jianran Liu, Weijian Sun, Tong Guo, Kechang Yang, Biao Hu, Yangang Zhang, Mingguo Zhao
Förderung
- Beijing Natural Science Foundation
- Science and Technology Innovation 2030 Major Project
- Tsinghua University Initiative Scientific Research Program
Zitieren
APA
Wang, Y., Luo, C., Chen, P., Liu, J., Sun, W., Guo, T., Yang, K., Hu, B., Zhang, Y., & Zhao, M. (2026). Learning vision-driven reactive soccer skills for humanoid robots. Science Robotics, 11(117). https://doi.org/10.1126/scirobotics.aed1152
Vancouver
Wang Y, Luo C, Chen P, Liu J, Sun W, Guo T, et al. Learning vision-driven reactive soccer skills for humanoid robots. Science Robotics. 2026;11(117). doi:10.1126/scirobotics.aed1152
BibTeX
@article{Wang2026,
author = {Wang, Yushi and Luo, Changsheng and Chen, Penghui and Liu, Jianran and Sun, Weijian and Guo, Tong and Yang, Kechang and Hu, Biao and Zhang, Yangang and Zhao, Mingguo},
title = {Learning vision-driven reactive soccer skills for humanoid robots},
journal = {Science Robotics},
year = {2026},
volume = {11},
number = {117},
doi = {10.1126/scirobotics.aed1152}
}
EUVIMED