Mobile embedded systems belong among the typical applications of distributed systems control in real-time. An example of a mobile control system is a robotic system. The proposal and realization of such a distributed control system represents a demanding and complex task for real-time control. In the process of robot soccer game applications, extensive data is accumulated. The reduction of such data is a possible win in a game strategy. The main topic of this article is a description of an efficient method for rule selection from a strategy. The proposed algorithm is based on the geometric representation of rules. A described problem and a proposed solution can be applied to other areas dealing with effective searching of rules in structures that also represent coordinates of the real world. Because this construed strategy describes a real space and the stores physical coordinates of real objects, our method can be used in strategic planning in the real world where we know the geographical positions of objects.