Visually-guided navigation by comparing edge images

Abstract

this paper we describe a method for using twodimensional shape information to determine the location of a mobile robot with respect to some visual landmark in the world. The task is for the robot to navigate to a specified target or landmark in its visual field, possibly in the presence of obstacles. The landmark is initially specified either by marking some portion of an image (containing the landmark) or by providing a prior model. The location of the landmark with respect to the robot is recovered from the change in apparent size and position of the landmark in the image, as a function of the motion of the robot. The size and position of the landmark are determined by comparing two-dimensional shapes from successive images taken as the robot moves.