Designing a skilled soccer team for RoboCup: exploring skill-set-primitives through reinforcement learning

TL;DR

This paper presents a reinforcement learning-based framework for developing a skilled humanoid soccer team in RoboCup, introducing novel skill primitives, a symmetry-extended PPO algorithm, and a comprehensive training pipeline that enhances skill efficiency and transition seamlessness.

Contribution

The paper introduces a new codebase, a set of skill primitives, and a symmetry-extended PPO algorithm, advancing autonomous soccer agent development and resource-efficient multi-agent learning.

Findings

Won RoboCup 2022 and 2023 competitions

Developed a multi-purpose omnidirectional walk and advanced ball control skills

Enhanced sample efficiency and skill transition stability

Abstract

The RoboCup 3D Soccer Simulation League serves as a competitive platform for showcasing innovation in autonomous humanoid robot agents through simulated soccer matches. Our team, FC Portugal, developed a new codebase from scratch in Python after RoboCup 2021. The team's performance relies on a set of skills centered around novel unifying primitives and a custom, symmetry-extended version of the Proximal Policy Optimization algorithm. Our methods have been thoroughly tested in official RoboCup matches, where FC Portugal has won the last two main competitions, in 2022 and 2023. This paper presents our training framework, as well as a timeline of skills developed using our skill-set-primitives, which considerably improve the sample efficiency and stability of skills, and motivate seamless transitions. We start with a significantly fast sprint-kick developed in 2021 and progress to the most recent skill set, including a multi-purpose omnidirectional walk, a dribble with unprecedented ball control, a solid kick, and a push skill. The push addresses low-level collision scenarios and high-level strategies to increase ball possession. We address the resource-intensive nature of this task through an innovative multi-agent learning approach. Finally, we release the team's codebase to the RoboCup community, providing other teams with a robust and modern foundation upon which they can build new features.