Mobile Robot Localization using Panoramic Vision and Combinations of Local Feature Region Detectors

This paper presents a vision-based approach for mobile robot localization. The environmental model is topological. The new approach uses a constellation of different types of local affine covariant regions to characterize a place. This type of representation permits a reliable and distinctive environment modeling. The goal of our work was to find out if using combinations of complementary local feature region detectors improves the localization with respect to using a single region detector. Our experimental results show that although the combination of different covariant affine region detectors increases the number of detected features and thus of potential matches it does not always improve the localization of the robot. Indeed, similarly to single detectors, different combinations also exhibit different strengths and weaknesses depending on the situation.