Journal of Computer & Robotics 6 (2), 2013 1-6 * Corresponding author. Email: keshavarzi.a@ut.ac.ir 1 Coordination Approach to Find Best Defense Decision with Multiple Possibilities among Robocup Soccer Simulation Team Ashkan Keshavarzi a* , Nader Zare b a Department of Electrical and Computer Engineering, University of Tehran, Tehran, Iran b Department of Computer Engineering, Khajeh-Nasir University, Tehran, Iran Received 7 March 2012; accepted 2 May 2012 Abstract In 2D Soccer Simulation league, agents will decide based on information and data in their model. Effective decisions need to have world model information without any noise and missing data; however, there are few solutions to omit noise in world model data; so we should find efficient ways to reduce the effect of noise when making decisions. In this article we evaluate some simple solutions when making defense decisions and try to find a solution based on message- passing to coordinating agents in defense situations. Our experimental results showed that in each situation one of the agents has a better view than others, so that agent can send messages to the others and provide needed information for doing defense behavior(ex: block behavior or clear ball behavior). Finally, we implement our solution based on Agent2D, version 3 and compare that with other solutions implemented in Cyrus2014 and Marlik2013 Soccer 2D simulation teams. Keywords: Multi-Agent coordination, Message-passing, Robocup soccer 2D simulation, Autonomous Agents, Decision Making 1. Introduction Robocup is improving artificial intelligence, robotic and others areas by presenting some standard problems in which different technologies are combined and tested.[1] In this article we try to use co-ordination between autonomous agents in soccer 2d simulation environment and implement better strategies for defense decision. In 2d soccer simulation league, which is a main branch of robotic science, determining game strategies are difficult because of the noise in the received world model data. Also in this environment agents are more restricted in their communication and calculation and they are less aware of the decisions of other agents; so in these years active teams in the competitions use an approach to teach their agents to find best decisions in such conditions. In different approaches introduced so far, one can mention the use of Kalman-filter to reduce noise in data, use offline methods in learning science, and the use of learning machine to choose the best behavior.[2] In this article we try to introduce new approaches to solve problems related to decision