rSoccer: A Framework for Studying Reinforcement Learning in Small and Very Small Size Robot Soccer

Reinforcement learning is an active research area with a vast number of applications in robotics, and the RoboCup competition interesting environment for studying evaluating reinforcement methods. A known difficulty applying to robotics high experience samples required, being use simulated environments training agents followed by transfer real-world (sim-to-real) viable path. This article introduces open-source simulator IEEE Very Small Size Soccer League optimized experiments. We also propose framework creating OpenAI Gym set benchmarks tasks single-agent multi-agent robot soccer skills. then demonstrate capabilities two state-of-the-art methods as well their limitations certain scenarios introduced this framework. believe will make it easier more teams compete these categories using end-to-end approaches further develop area.