Pounds D scene modeling for autonomous mobile robot self-location using passive landmarks

Considers the problems of finding the location and orientation of a mobile robot using an image taken by a CCD camera carried by the robot. The naturally occurring landmarks used are the corners of the room extracted from the line image of the 3D indoor scene. The location and orientation of the vehicle is calculated by perspective information of the landmarks in the scene of the room where the robot moves. This system and some experiments to demonstrate the performance of its components are described.<
>