English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The RoboCup 3D Soccer Simulation League serves as a competitive platform forshowcasing innovation in autonomous humanoid robot agents through simulatedsoccer matches. Our team, FC Portugal, developed a new codebase from scratch inPython after RoboCup 2021. The team's performance is based on a set of skillscentered around novel unifying primitives and a custom, symmetry-extendedversion of the Proximal Policy Optimization algorithm. Our methods have beenthoroughly tested in official RoboCup matches, where FC Portugal has won thelast two main competitions, in 2022 and 2023. This paper presents our trainingframework, as well as a timeline of skills developed using ourskill-set-primitives, which considerably improve the sample efficiency andstability of skills, and motivate seamless transitions. We start with asignificantly fast sprint-kick developed in 2021 and progress to the mostrecent skill set, which includes a multi-purpose omnidirectional walk, adribble with unprecedented ball control, a solid kick, and a push skill. Thepush tackles both low-level collision-prone scenarios and high-level strategiesto increase ball possession. We address the resource-intensive nature of thistask through an innovative multi-agent learning approach. Finally, we releasethe codebase of our team to the RoboCup community, enabling other teams totransition to Python more easily and providing new teams with a robust andmodern foundation upon which they can build new features.

Designing a Skilled Soccer Team for RoboCup: Exploring Skill-Set-Primitives through Reinforcement Learning