A Stereo Vision-Based Self-Localization System

This paper proposes a new stereo vision-based self-localization system (SVBSLS) for the RoboCup soccer humanoid league rules for the 2010 competition. The humanoid robot integrates information from pan/tilt motors and stereo vision to accomplish the self-localization and measure the distance from the robot to the soccer ball. The proposed approach uses a trigonometric function to find coarse distances from the robot to the landmark and from the robot to the soccer ball, after which it adopts an artificial neural network technique to increase the precision of the distance measurement. A statistics approach is also used to calculate the relationship between the humanoid robot and the position of the landmark for self-localization. The experimental results indicate that the SVBSLS localization system in this paper had an accuracy ratio of 100 for localization. The average error of distance from the humanoid soccer robot to the soccer ball was only 0.64 cm.