Centralized Robot Soccer Architecture Based on Roles

被引:1
作者
Guarnizo, Jose G. [1 ,2 ]
Mellado, Martin [2 ]
机构
[1] Univ Dist Francisco Jose de Caldas, Lab Invest Fuentes Alternat Energia, Carrera 7 40-53 Piso 5, Bogota, Colombia
[2] Univ Politecn Valencia, Inst Automat & Informat Ind, Camino Vera S-N, E-46022 Valencia, Spain
来源
REVISTA IBEROAMERICANA DE AUTOMATICA E INFORMATICA INDUSTRIAL | 2016年 / 13卷 / 03期
关键词
Agents; decision making; autonomous mobile robots; centralized control; architectures; STRATEGY; COORDINATION; ALLOCATION; MECHANISM; SELECTION;
D O I
10.1016/j.riai.2016.05.005
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Robot soccer offers an adequate domain in order to design and validate architectures for robot-coordination. One classification refers to centralized architectures, which correspond to robot soccer environments with global perception and centralized control of the robots, using only one decision-making system. In this paper it is presented a centralized robot soccer architecture based on roles, where one role is assigned to each player in order to select a specific behaviour depending on game conditions. Roles are assigned using an assignment function, which is activated when the ball changes of the quadrant in the playing field. This strategy has been compared by simulation in games against an opposition team with constant roles, and other team with a hierarchical strategy which assigns roles depending on a tactic previously selected. The results showed a better performance in the team with the role-based strategy outperformed the rest of the methods. As well as uniformity within the players' behaviors during the role and behavior transitions.
引用
收藏
页码:370 / 380
页数:11
相关论文
共 46 条
[1]  
Abeyruwan S., 2012, 7 WORKSH HUM SOCC RO
[2]   Improving a simulated soccer team's performance through a Memory-Based Collaborative Filtering approach [J].
Abreu, Pedro Henriques ;
Silva, Daniel Castro ;
Almeida, Fernando ;
Mendes-Moreira, Joao .
APPLIED SOFT COMPUTING, 2014, 23 :180-193
[3]  
Acosta Calderon C. A., 2010, ROBOT SOCCER CHAPTER, P121, DOI [10.5772/7344, DOI 10.5772/7344]
[4]  
Aguero C. E., 2006, 2 INT WORKSH MULT RO
[5]   On-the-fly generation of multi-robot team formation strategies based on game conditions [J].
Atkinson, John ;
Rojas, Dario .
EXPERT SYSTEMS WITH APPLICATIONS, 2009, 36 (03) :6082-6090
[6]  
Baldoni M., 2008, WORKSH OBJ AG, P89
[7]   A review of swarm robotics tasks [J].
Bayindir, Levent .
NEUROCOMPUTING, 2016, 172 :292-321
[8]   Distributed Artificial Intelligence based Architecture applied to the Integrated Industrial Production Management [J].
Bravo, Cesar ;
Aguilar-Castro, Jose ;
Rios, Addison ;
Aguilar-Martin, Joseph ;
Rivas, Francklin .
REVISTA IBEROAMERICANA DE AUTOMATICA E INFORMATICA INDUSTRIAL, 2011, 8 (04) :405-417
[9]   Multi-agent role allocation: issues, approaches, and multiple perspectives [J].
Campbell, Adam ;
Wu, Annie S. .
AUTONOMOUS AGENTS AND MULTI-AGENT SYSTEMS, 2011, 22 (02) :317-355
[10]  
Cardoso P., 2012, 2012 Brazilian Robotics Symposium and Latin American Robotics Symposium (SBR-LARS 2012), P143, DOI 10.1109/SBR-LARS.2012.30