Learning Soccer Skills for Humanoid Robots: A Progressive Perception-Action Framework

TL;DR

This paper introduces PAiD, a progressive perception-action framework for humanoid robots to learn soccer skills through staged training, resulting in stable, robust, and adaptable kicking performance in diverse real-world conditions.

Contribution

The paper presents a novel staged learning architecture that decomposes soccer skill acquisition into three stages, improving stability, perception integration, and sim-to-real transfer for humanoid robots.

Findings

Achieved high-fidelity, human-like kicking on Unitree G1 robots.

Demonstrated robustness under diverse conditions and disturbances.

Maintained consistent performance across indoor and outdoor environments.

Abstract

Soccer presents a significant challenge for humanoid robots, demanding tightly integrated perception-action capabilities for tasks like perception-guided kicking and whole-body balance control. Existing approaches suffer from inter-module instability in modular pipelines or conflicting training objectives in end-to-end frameworks. We propose Perception-Action integrated Decision-making (PAiD), a progressive architecture that decomposes soccer skill acquisition into three stages: motion-skill acquisition via human motion tracking, lightweight perception-action integration for positional generalization, and physics-aware sim-to-real transfer. This staged decomposition establishes stable foundational skills, avoids reward conflicts during perception integration, and minimizes sim-to-real gaps. Experiments on the Unitree G1 demonstrate high-fidelity human-like kicking with robust performance under diverse conditions-including static or rolling balls, various positions, and disturbances-while maintaining consistent execution across indoor and outdoor scenarios. Our divide-and-conquer strategy advances robust humanoid soccer capabilities and offers a scalable framework for complex embodied skill acquisition. The project page is available at https://soccer-humanoid.github.io/.