Skip to main content

RoboCup 2024

Link to all files

YuShan2024

  • 📄 Title: YuShan2024 Team Description Paper for RoboCup2024
  • ✍️ Authors: Zekai Cheng, Liang Zhang, Dabo Dan, Huiyu Xiong, Haiyang Qin
  • 📜 Abstract: This team description paper describes the direction and methods of team optimization over the past year at YuShan2024. There are two main parts of the work, which are the analysis and optimization of the shooting module and the analysis of physical strength and running distance. In a lot of tests in the past, we found that the team has the situation that even though the player is in front of the goal and has a chance to shoot, but there is no shooting action, and missed some good chances, so we hope to find the main direction to optimize the shooting module by analyzing the shooting data from multiple perspectives. At the same time, in the test with some strong teams, we found that YuShan's stamina maintenance has a big problem, often around 500 cycles before the end of the half, the player's stamina has been exhausted, resulting in the defense out of position, so we hope that through the method of data analysis, we can find the difference in the stamina maintenance of the players between us and the strongest teams, so as to further make optimization of the team's stamina maintenanc.

AEteam

  • 📄 Title: AEteam Soccer Simulation 2D Team Description Paper 2024
  • ✍️ Authors: Erfan Fathi , Soroush Mazloom , Parham Keyhani , Amir hosein Nikfetrat and Vahid khodabakhshi
  • 📜 Abstract: This team description paper presents an overview of previous work and recent research topics of Team AEteam. In this article, we aim to enhance our understanding and improve the efficiency of offensive players escaping by exploring and expanding our knowledge on the subject. To achieve this, we em- ploy the mentioned artificial intelligence methods with the hope of improving the trend of players’ attacks.

CYRUS2024

  • 📄 Title: Observation Denoising in CYRUS Soccer Simulation 2D Team For RoboCup 2024
  • ✍️ Authors: Nader Zare, Aref Sayareh, Sadra Khanjari, and Arad Firouzkouhi
  • 📜 Abstract: In the Soccer Simulation 2D environment, accurate observa- tion is crucial for effective decision-making. However, challenges such as partial observation and noisy data can hinder performance. To address these issues, we propose a denoising algorithm that leverages predictive modeling and intersection analysis to enhance the accuracy of observa- tions. Our approach aims to mitigate the impact of noise and partial data, leading to improved gameplay performance. This paper presents the framework, implementation, and preliminary results of our algorithm, demonstrating its potential in refining observations in Soccer Simulation 2D. Cyrus 2D Team is using a combination of Helios, Gliders, and Cyrus base codes.

HELIOS2024

  • 📄 Title: HELIOS2024: Team Description Paper
  • ✍️ Authors: Hidehisa Akiyama, Tomoharu Nakashima, Kyo Hatakeyama, Takumi Fujikawa, and Akei Hishiki
  • 📜 Abstract: This team description paper presents an overview of previous work and recent research topics of Team HELIOS2024. This year, we en- hanced the online opponent formation identification, which our team had previously implemented. The primary contribution involves accelerating computations through the introduction of tile coding. This paper describes one of our recent efforts, the enhancement of oppo- nent identification.

Mars

  • 📄 Title: MARS Soccer Simulation 2D Team Description Paper 2024
  • ✍️ Authors: Mehdi torshani , Mohammad Maziani , Mohammad Ghasemi , Mohammad Bonakdar Tehrani , Fakhteh Hosseinkhani , Mahdieh GhasemPour , Afsaneh Imani , Zahra Samadi ,and Mercedeh Ghasemi
  • 📜 Abstract: Report examines the records and performance of the Mars team in the two-dimensional football simulation game.The Mars team began its activities in robotics in 2006. This team has developed various algorithms in the ation league using the Agent2D base code. In this report, we will discuss part of this team's goals in decision-making methods and the player learning system. The algorithm discussed in this report is a decision tree that makes the best decision for each player using the highest probability and the best training data.

Oxsy

  • 📄 Title: OXSY 2024 Team Description
  • ✍️ Authors: Sebastian Marian, Dorin Luca, Radu Sacuiu, Bogdan Sarac, Ovidiu Cotarlea
  • 📜 Abstract: Oxsy team has been founded in July 2002 as a graduation project of one student, Sebastian Marian, in the field of Multi-Agent Systems [1] Lucian Blaga University (Sibiu - Romania). After graduation he continued the work on this project and eventually Oxsy team was born. As we started from scratch [2] our ideas, concepts and beliefs, have been implemented year by year, and today, we are happy to see our evolution, as our team was growing in these years, much more than we have expected from the beginning. If we will qualify for this year competition, we will reach at the 20th consecutive participation, in RoboCup [3] Soccer Simulation League.

RobôCIn

  • 📄 Title: RobˆoCIn Soccer Simulation 2D Team Description Paper for RoboCup 2024
  • ✍️ Authors: Bonna Borsoi, Felipe Pereira, Gabriel Souza, Heitor Souza, Jeferson Ara ́ujo, Maria L. Silva, Mariana Coelho, Mateus Machado, Mateus Soares, Pedro V. Cunha, Rafael Labio, Walber M. Rodrigues, and Edna Barros
  • 📜 Abstract: RobˆoCIn Soccer Simulation 2D team, based at the Universi- dade Federal de Pernambuco, was founded in 2018. In our debut compe- tition at the Latin American Robotics Competition (LARC) in Jo ̃ao Pes- soa, Para ́ıba, Brazil, we secured fourth place against other Latin Amer- ican teams. The following year, we competed in the RoboCup for the first time and achieved a ninth-place finish. In 2023, we placed in the 6th position on RoboCup and first place at LARC. In this paper, we present the work developed over the past year, especially the refinement and improvement of the agent’s behaviors and changes made to the field evaluator.

FRA-UNIted2024

  • 📄 Title: FRA-UNIted — Team Description 2024
  • ✍️ Authors: Thomas Gabel, Berkan Eren, Eicke Godehardt
  • 📜 Abstract: The main focus of FRA-UNIted’s effort in the RoboCup soc- cer simulation 2D domain is to develop and to apply machine learning techniques in complex domains. In particular, we are interested in ap- plying reinforcement learning methods, where the training signal is only given in terms of success or failure. In this paper, we review some of our recent efforts taken during the past year, putting a special focus on tool support and team performance analyses.

ITAndroids

  • 📄 Title: ITAndroids 2D Soccer Simulation Team Description Paper 2024
  • ✍️ Authors: Davi M. Vasconcelos, Luiz F. B. Ramos, Marcos R. O. A. Maximo, Nean Segura, and Vinícius F. Almeida
  • 📜 Abstract: The ITAndroids 2D Soccer Simulation team is composed of undergraduate students of the Aeronautics Institute of Technology. This paper explores three topics: defensive behavior for 1-versus-1 situations with Deep Reinforcement Learning, improvement of the goalkeeper’s po- sitioning, and influence of the librcsc’s version on the agent2d perfor- mance. The developed defensive behavior outperformed agent2d in the presented task. The novel goalkeeper’s implementation dramatically de- creased the loss rate of ITAndroids against RoboCIn. Furthermore, the agent2d performance considerably depended on the librcsc’s version: the latest agent2d running with the most updated librcsc release achieved a win rate of 71.6% against agent2d 3.1.0 with librcsc 4.1.0.

R2D2

  • 📄 Title: R2D2 Soccer Simulation 2D Team Description Paper Robocup World Cup 2024
  • ✍️ Authors: Seyed Hassan Majid Zonouzi, Mohammad Hesam Nasiri, Sanaz Moosapour, Sina Taheri Behrooz, Sana Mousavi, Seyed Mostafa Atyabi, Fatemeh Houra Haghighatkhah
  • 📜 Abstract: The paper details the R2D2 team’s (R3CESBU) algorithm and de- velopment clarifications for RoboCup's 2D soccer simulation league. It intro- duces goalie, shooting, and marking advancements, emphasizing the ranking al- gorithm for chain action states and dynamic positioning. R2D2 employs artificial intelligence techniques, combining behavioral cloning and game log parsing to enhance kickable agents and goalies. The research contributes to the evolution of competitive play in the RoboCup 2D Simulation League. The offence tactics in- volve improvements in the field evaluator, through-pass, unmarking, and shoot- ing algorithms, resulting in significant performance enhancements.