Visual robot detection in robocup using neural networks

Authors:
Ulrich Kaufmann;Gerd Mayer;Gerhard Kraetzschmar;Günther Palm
Affiliations:
Department of Neural Information Processing, University of Ulm, Ulm, Germany;Department of Neural Information Processing, University of Ulm, Ulm, Germany;Department of Neural Information Processing, University of Ulm, Ulm, Germany;Department of Neural Information Processing, University of Ulm, Ulm, Germany
Venue:
RoboCup 2004
Year:
2005

Citing 6
Cited 3

Artificial intelligence: a modern approach

Artificial intelligence: a modern approach
Fast and Accurate Robot Vision for Vision Based Motion

RoboCup 2000: Robot Soccer World Cup IV
A Fast Vision System for Middle Size Robots in RoboCup

RoboCup 2001: Robot Soccer World Cup V
Robust Real Time Color Tracking

RoboCup 2000: Robot Soccer World Cup IV
Object Classification Using Simple, Colour Based Visual Attention and a Hierarchical Neural Network for Neuro-symbolic Integration

KI '99 Proceedings of the 23rd Annual German Conference on Artificial Intelligence: Advances in Artificial Intelligence
Watch their moves: applying probabilistic multiple object tracking to autonomous robot soccer

Eighteenth national conference on Artificial intelligence

Reinforcement learning for robot soccer

Autonomous Robots
Optical Flow Based Detection in Mixed Human Robot Environments

ISVC '09 Proceedings of the 5th International Symposium on Advances in Visual Computing: Part I
Neural robot detection in robocup

Biomimetic Neural Learning for Intelligent Robots

Quantified Score

Hi-index	0.00

Visualization

Abstract

Robot recognition is a very important point for further improvements in game-play in RoboCup middle size league. In this paper we present a neural recognition method we developed to find robots using different visual information. Two algorithms are introduced to detect possible robot areas in an image and a subsequent recognition method with two combined multi-layer perceptrons is used to classify this areas regarding different features. The presented results indicate a very good overall performance of this approach.