A tool which can be used for the analysis of a robot soccer game is presented. The tool enables automatic filtering and selection of game sequences which are suitable for the analysis of the game. Fuzzy logic is used since the data gathered by a camera is highly noisy. The data used in the paper was recorded during the game Germany - Slovenia in Hagen, on No- vember 11, 2001. The dynamic parameters of our robots are estimated using the least squares technique. Meandering parameters are estimated and an attempt is made to identify the strategy of the opposing team with the method of introspection. Robot soccer has recently become popular all over the world. It is used as a test bed for various technical solutions and scientific methods. In competition sys- tems, such as robot soccer, the analysis of a game is useful for the detection of faults in the program and enables the applied algorithms to be improved. The analysis is performed based on data recorded during the game. The output of the vision system is recorded for subsequent analysis. Most robot soccer programs consist of two main parts, both of which may mal- function: the vision and strategy components. In this paper we will focus on the analysis of the strategy, which is done "manually" with the help of an off line computer aided interactive tool. Performing such an analysis is important from the competitive point of view and may be helpful for the elimination of bugs and faults in own strategy as well as for the detection of weak points in the opposing team and for the preparation of the strategy to be used in forthcoming duels. The tool employed should automatically perform the identification of critical situations for off line analysis. This paper deals with the application of identification methods to the detection of critical situations which may be of interest for the subsequent analysis. Actually the identified parameters are kine- matic parameters, but they are closely connected with the applied strategy; e.g. the time needed to kick the
[1]
Henk B. Verbruggen,et al.
Demystification of Fuzzy Control
,
1994
.
[2]
Seong-Wook Park.
System Development for Education and Design of a Nonlinear Controller with On-Line Algorithm
,
2003
.
[3]
Xiao-Jun Zeng,et al.
Approximation theory of fuzzy systems-SISO case
,
1994,
IEEE Trans. Fuzzy Syst..
[4]
Lennart Ljung,et al.
Nonlinear black-box modeling in system identification: a unified overview
,
1995,
Autom..
[5]
Robert Babuška,et al.
An overview of fuzzy modeling for control
,
1996
.
[6]
Michio Sugeno,et al.
Fuzzy identification of systems and its applications to modeling and control
,
1985,
IEEE Transactions on Systems, Man, and Cybernetics.
[7]
Xiao-Jun Zeng,et al.
Approximation theory of fuzzy systems-MIMO case
,
1995,
IEEE Trans. Fuzzy Syst..