Vision-Based Dribbling for Humanoid Soccer via Privileged Representation Learning

2026-07-14

AI Summary

This work addresses the challenge of enabling humanoid robots to perform dribbling maneuvers in dynamic adversarial environments using only onboard vision. The authors propose an end-to-end reinforcement learning framework that integrates perception and control, embedding a temporal depth image encoder and a task-specific projection layer directly into the policy network. This architecture learns dribbling policies directly from raw visual inputs without relying on explicit state estimation or privileged information. Evaluated on the Booster T1 humanoid robot in simulation, the method achieves a 100% success rate in goal-oriented dribbling tasks, 96% in scenarios with static obstacles, and 46% against actively defending opponents, demonstrating substantially improved robustness and effectiveness under occlusions, rapid ball motion, and complex physical interactions.

Abstract

Recent advances in humanoid robotics have highlighted the importance of deployable loco-manipulation skills. Dribbling a soccer ball while evading active opponents requires simultaneous balance, precise ball control, and awareness of a dynamic adversary under onboard sensing and real-time constraints. Existing approaches typically separate perception and motion, which can be effective in controlled settings but may fail under occlusions, fast ball movements, and complex opponent interactions, since perception is not directly optimized for control. We propose an integrated approach in which a temporal depth encoder is embedded into a reinforcement learning policy through a task-specific projection layer. We apply this framework to a simulated Booster T1 humanoid robot and show that it is possible to learn vision-based, opponent-aware dribbling directly from depth observations, without explicit state estimation or privileged scene information. The learned policy achieves 100% success in nominal target-driven dribbling and 96% success with a single static obstacle, while reaching 46% success against an actively moving ball-attacker opponent. These results demonstrate that the proposed framework supports robust vision-based dribbling in nominal and moderately dynamic settings, and provides a strong foundation for handling more challenging moving-adversary scenarios.

Problem

• vision-based dribbling
• humanoid soccer
• loco-manipulation
• dynamic opponent
• onboard sensing

Innovation

• vision-based control
• privileged representation learning
• humanoid soccer
• loco-manipulation
• reinforcement learning