Robust Self-Localization using Elastic Templates

Abstract

A visually orienting mobile robot must cope with a number of changes to its environment. Most importantly, it must be able to identify its location even when objects in the environment have been moved or when the illumination conditions have changed. We report experiments with a self-localization method based on elastic templates that operates in real time on a Pentium 133 MHz PC. The algorithm has been implemented in an autonomous mobile robot as an important part of its navigational system. We investigate how well the algorithm generalizes between different locations and angles toward the landmark when the robot is moved around. We also report experiments with different illumination and partial occlusion of the landmarks. We show that elastic template matching is very robust against changing environments.