Self-localization of mobile robot with single camera in corridor environment

This paper presents a single vision-based self-localization method in an corridor environment. The authors use the Hough transform for finding parallel lines and vertical lines. Relative distance from mobile robot to the cross points is calculated by using the cross points as feature points. Mapping an environment map to the feature points is performed by self-localization in a searching window. Experimental results show the efficiency of the proposed method in a corridor environment.