Stereo AoA system for indoor SLAM

摘要

In this paper, we address two systems composed of ODVS(Omni-directional Vision Sensor) for localizing mobile robots in indoor environments. One, with single AoA(Angle of Arrival) system, needs at least three landmarks of predefined location while the other, called stereo AoA system, requires only one landmark of known location for localization. For identifying the landmarks, image processing techniques are applied to omni-directional image to get colored information that is main features of indoor environments. The final goal of this research is to build up local map autonomously by image processing in indoor environments using lines on the image sequences. Through several experiments with the system, we confirmed the feasibility of the system for map building.