Abstract
This paper presents a method of mobile robot's autonomous navigation in indoor environments by a teaching and playing-back scheme using memories of an ounri-directional image sequence and robot's motion. The robot memorizes a sequence of environmental images and robot's poses during a teaching stage. And in the course of a playing-back stage. itt calculates the robot's position difference from the memorized and currently taken images, and then decides a trajectory to track the taught route, so that it can navigate autonomously to follow the given path to the destination. An omni-directional image sensor is used, and vertical edges arc extracted as feature lines in the environment for calculation of robot's position difference. The experimental results show the effectiveness of this method in corridors of bulidings.
