A FastSLAM-based algorithm for omnidirectional cameras

  1. Gamallo Solórzano, Cristina
  2. Mucientes Molina, Manuel
  3. Regueiro, Carlos V.
Revista:
JoPha: Journal of Physical Agents

ISSN: 1888-0258

Ano de publicación: 2013

Título do exemplar: Special issue on advances on physical agents

Volume: 7

Número: 1

Páxinas: 3

Tipo: Artigo

DOI: 10.14198/JOPHA.2013.7.1.03 DIALNET GOOGLE SCHOLAR lock_openRUA editor

Outras publicacións en: JoPha: Journal of Physical Agents

Resumo

Environments with a low density of landmarks are difficult for vision-based Simultaneous Localization and M apping (SLAM) algorithms. The use of omnidirectional cameras, whi ch have a wide field of view, is specially interesting in these environments as several landmarks are usually detected in e ach image. A typical example of this kind of situation happens in indoor environments when the lights placed on the ceiling ar e the landmarks. The use of omnivision combined with this type of landmarks presents two challenges: the data association and the initialization of the landmarks with a bearing-only sen sor. In this paper we present a SLAM algorithm based on the well- known FastSLAM approach [1]. The proposal includes a novel hierarchical data association method based on the Hungaria n algorithm, and a delayed initialization of the landmarks. T he approach has been tested on a real environment with a Pioneer 3-DX robot.