Dribble Master: Learning Agile Humanoid Dribbling through Legged Locomotion

TL;DR

This paper presents a two-stage curriculum learning approach for humanoid robots to learn agile soccer dribbling, combining simulation training with real-world transfer, and emphasizing active perception for effective ball control.

Contribution

Introduces a novel two-stage curriculum learning framework with a virtual camera model and heuristic rewards for realistic and adaptable humanoid dribbling skills.

Findings

Successful transfer of policies from simulation to real robot

Effective ball manipulation and agile dribbling behaviors achieved

Enhanced perception through active sensing strategies

Abstract

Humanoid soccer dribbling is a highly challenging task that demands dexterous ball manipulation while maintaining dynamic balance. Traditional rule-based methods often struggle to achieve accurate ball control due to their reliance on fixed walking patterns and limited adaptability to real-time ball dynamics. To address these challenges, we propose a two-stage curriculum learning framework that enables a humanoid robot to acquire dribbling skills without explicit dynamics or predefined trajectories. In the first stage, the robot learns basic locomotion skills; in the second stage, we fine-tune the policy for agile dribbling maneuvers. We further introduce a virtual camera model in simulation that simulates the field of view and perception constraints of the real robot, enabling realistic ball perception during training. We also design heuristic rewards to encourage active sensing, promoting a broader visual range for continuous ball perception. The policy is trained in simulation and successfully transferred to a physical humanoid robot. Experiment results demonstrate that our method enables effective ball manipulation, achieving flexible and visually appealing dribbling behaviors across multiple environments. This work highlights the potential of reinforcement learning in developing agile humanoid soccer robots. Additional details and videos are available at https://zhuoheng0910.github.io/dribble-master/.