Object Recognition and Self-Localization for Interactive Soccer Robots

This paper presents an experimental study on object recognition and self-localization for multiagent robotic soccer. A color image segmentation method based on YUV color information of each pixel of the color image is applied to the ball recognition and self-localization of a small mobile robot. First the lower and higher threshold values of Y, U, and V are determined for the sample color markers. Then color image segmentation is performed using the tables that transform each value of Y, U, and V in a pixel to membership in several color classes. The position of objects on the field is efficiently and accurately calculated using a monocular camera. Self-localization is accomplished using the distances and orientations of two landmarks. Experimental results of real-time movement control are also presented.