Formation Designing And Passing Strategy In Simulation 2D Soccer System

RoboCup(Robot World Cup)2D simulation platform is a dynamic multi-agent competition system. Agents need to do rational decision-making within a limited time, compete with the opposite team in the soccer field. Similar to human soccer, dynamic agent positioning and flexible passing strategy are the key points to win the game. Therefore, our research focuses on formation designing, passing strategy. The main work and innovations are as follows:First of all, in this paper, we use Delaunay triangulation to realize our team formation. This method is a kind of SBSP(Situation Based Strategy Positioning). We transform the problem into a map from a focal point like a ball position in a soccer field to a ideal coordinates of each player agent, and split the soccer field into a set of triangles using Delaunay triangulation to approximate the map. With the method, our agents can realize flexible positioning according to the ball movement.Secondly, we form our passing strategy with A~* search. In this part, passing strategy problem is transformed into a ball path planning problem that we find the optimal way to let the ball travel from the start point to the goal point under the obstacles environment. The method we use to find the path is the A~* search. At the same time, we use Newton iteration method to optimize the calculation of the interception process. So that we provide optimal interception to the upper passing strategyAt the end of this paper, we combined with the above two points to design the interception formation and the offensive formation. Experimental and competition results certify the validity and feasibility of our methods.