Integrating Planning and a Rule-Based System for Controlling a Robot Soccer Team

There are many potential field techniques to control multiple robots. Among them are Arkin´s Potential Field, Harmonic Potential Field and other ones based on the solution of the Boundary Value Problem (BVP). Most of them have been applied to control just one robot. In this paper, a modification in a technique of Potential Field based on the solution of BVP is proposed for controlling multiple robots. This modification is called Locally Oriented Potential Field (LOPF) and it overcomes the drawback of BVP based techniques when they are applied for controlling multi robots. Besides, a rule based system has been proposed to characterize the functionality of each robot for realizing an specific task. This is necessary for that the behavior of the equip members not be competitive but collaborative. It will be shown that LOPF is very appropriated to control robot soccer teams, in which each robot has the necessity of adapting its behavior during a game. Furthermore, some experiment results are presented to show the efficiency of the proposed technique.