Achieving accurate mobile robot localization by combining topological maps and raw sonar readings

• Autores: Antonio Burguera Burguera, Yolanda González Cid, Gabriel Oliver Codina
• Localización: XXVI Jornadas de Automática: 7-10 de septiembre de 2005, Alicante-Elche / coord. por Fernando Torres Medina, Óscar Reinoso García, 2005, ISBN 84-689-0730-8, págs. 639-646
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • This paper presents a mobile robot selflocalization system that takes advantage of both geometrically constrained and raw reading based methods using sonar. This system is structured in three layers. The lower layer operates at high frequency, providing the robot continuously an estimation of its position. This is accomplished by simply doing dead reckoning. The intermiediate layer operates at a lower frequency and uses the raw range information provided by sonar sensors to perform scan matching. Finally, the higher layer operates sporadically by exploiting the geometry of indoor environments through a topological map. Thanks to these three layers an accurate localization is achieved, as demonstrated by the experimental results.