Monte Carlo Localization using the Global Appearance of Omnidirectional Images. Algorithm Optimization to Large Indoor Environments
Lorenzo Fernández, Luis Payá, David Valiente, Arturo Gil and Oscar Reinoso
ICINCO 2012  (Rome, Italy, 28-31 July, 2012)
Ed. Copyright 2012 SciTePress – Science and Technology Publications All rights reserved  ISBN:978-989-8565-22-8  - 9th International Conference on Informatics in Control, Automation and Robotics


In this paper we deal with the problem of robot localization using the visual information provided by a single omnidirectional camera mounted on the robot, using techniques based on the global appearance of panoramic images. Our main objective consists in showing the feasibility of the appearance-based approaches in a localization task in a relatively large and real environment. First, we study the approaches that permit us to describe globally the visual information so that it represents with accuracy locations in the environment. Then, we present the probabilistic approach we have used to compute the most probable pose of the robot when it performs a trajectory within the map. At the end, we describe the kind of environments and maps we have used to test our localization algorithms and the final results. The experimental results we show have been obtained using real indoor omnidirectional images, captured in an office building under real conditions.