Neural Robot Detection in RoboCup

Improving the game play in RoboCup middle size league requires a very robust visual opponent and teammate detection system. Because RoboCup games are highly dynamic, the detection system also has to be very fast. That both conditions are not necessarily contradictory is shown in this paper. The described multilevel approach documents, that the combination of a simple color based attention control and a subsequent neural object classification can be applied successfully in real world scenarios. The presented results indicate a very good overall performance regarding robustness, flexibility and computational needs.

[1]  Jurjen Caarls,et al.  Fast and Accurate Robot Vision for Vision Based Motion , 2000, RoboCup.

[2]  Wolfram Burgard,et al.  KI-99: Advances in Artificial Intelligence , 2003, Lecture Notes in Computer Science.

[3]  Andreas Zell,et al.  Real-time object tracking for soccer-robots without color information , 2004, Robotics Auton. Syst..

[4]  Michael Beetz,et al.  Toward RoboCup without Color Labeling , 2002, AI Mag..

[5]  Vahab S. Mirrokni,et al.  A Fast Vision System for Middle Size Robots in RoboCup , 2001, RoboCup.

[6]  Thomas Röfer,et al.  Realtime Object Recognition Using Decision Tree Learning , 2004, RoboCup.

[7]  Günther Palm,et al.  Object classification with simple visual attention and a hierarchical neural network for subsymbolic-symbolic coupling , 1999, Proceedings 1999 IEEE International Symposium on Computational Intelligence in Robotics and Automation. CIRA'99 (Cat. No.99EX375).

[8]  Günther Palm,et al.  Concurrent Object Identification and Localization for a Mobile Robot , 2000, Künstliche Intell..

[9]  Hans Utz,et al.  Towards autonomous vision self-calibration for soccer robots , 2002, IEEE/RSJ International Conference on Intelligent Robots and Systems.

[10]  Michael Beetz,et al.  Toward RoboCup without color labeling , 2003 .

[11]  Hans Utz,et al.  Playing Robot Soccer under Natural Light: A Case Study , 2003, RoboCup.

[12]  Friedhelm Schwenker,et al.  Learning Object Recognition in a NeuroBotic System , 2004 .

[13]  Günther Palm,et al.  Visual Robot Detection in RoboCup Using Neural Networks , 2004, RoboCup.

[14]  Sven Behnke,et al.  Robust Real Time Color Tracking , 2000, RoboCup.

[15]  Günther Palm,et al.  Object Classification Using Simple, Colour Based Visual Attention and a Hierarchical Neural Network for Neuro-symbolic Integration , 1999, KI.

[16]  Javier Ruiz-del-Solar,et al.  Evolving Visual Object Recognition for Legged Robots , 2003, RoboCup.