A remarkable feature of RoboCup’s soccer simulation leagues is their ability to quantify and prove the exact progress made over years. In this paper, we present and discuss the results of an extensive empirical study of the progress and the currently reached state of 2D soccer simulation. Our main finding is that the current decade has witnessed a continuous and statistically significant improv...

In the Markov Decision Process (MDP) formalization of Reinforcement Learning, a single adaptive agent interacts with the environment defined by a probabilistic transition function. Secondary agents can only be a part of the environment and are therefore fixed in their behavior. The framework of Markov games allows us to widen this view to include multiple adaptive agents with interacting or com...

This paper presents a system for marking or covering players on an opposing soccer team so as to best prevent them from scoring. A basis for the marking system is the introduction of prioritized role assignment, an extension to SCRAM dynamic role assignment used by the UT Austin Villa RoboCup 3D simulation team for formational positioning. The marking system is designed to allow for decentraliz...

Physically-realistic simulated environments are powerful platforms for enabling measurable, replicable and statistically-robust investigation of complex robotic systems. Such environments are epitomised by the RoboCup simulation leagues, which have been successfully utilised to conduct massivelyparallel experiments in topics including: optimisation of bipedal locomotion, self-localisation from ...

The RoboCup simulated soccer league is a dynamic, complex and uncertain environment which presents many challenges to machine learning techniques. The asynchronous design of the RoboCup simulation environment can create long and unpredictable delays in the effects of actions, often causing onerous training times. A new environment known as Simple Soccer is proposed which, while retaining much o...

The main focus of the Brainstormers’ effort in the RoboCup soccer simulation 2D domain is to develop and to apply machine learning techniques in complex domains. In particular, we are interested in applying reinforcement learning methods, where the training signal is only given in terms of success or failure. Our final goal is a learning system, where we only plug in “win the match” – and our a...

Simulators have a long tradition within RoboCup. In the Soccer Simulation League, one of the earliest leagues within RoboCup, traditionally single-purpose robot simulators with high levels of abstraction have been used. These systems proved valuable as tools for multiagent research, but were difficult to extend. Furthermore, there were concerns that research results would not be easily transfer...

In the domain of RoboCup 2D soccer simulation league, appropriate player positioning against a given opponent team is an important factor of soccer team performance. This work proposes a model which decides the strategy that should be applied regarding a particular opponent team. This task can be realized by applying preliminary a learning phase where the model determines the most effective str...

Soccer simulation is an effort to motivate researchers and practitioners to do artificial and robotic intelligence research; and at the same time put into practice and test the results. Many researchers and practitioners throughout the world are continuously working to polish their ideas and improve their implemented systems. At the same time, new groups are forming and they bring bright new th...

This paper presents the team coordination methodologies of CAMBADA, a robotic soccer team designed to participate in the RoboCup middle-size league (MSL). The coordination model extends and adapts previous work in the Soccer Simulation League to the MSL environment. The approach is based on flexible positionings and priority-based dynamic role/positioning assignment. In addition, coordinated pr...