Unmanned Aerial Vehicle Localization Based on Monocular Vision and Online Mosaicking

• Fernando Caballero ^[1] ; Luis Merino ^[2] ; Joaquín Ferruz Melero ^[1] ; Aníbal Ollero Baturone ^[1]
  1. [1] Universidad de Sevilla

     Universidad de Sevilla

     Sevilla, España

     
  2. [2] Universidad Pablo de Olavide

     Universidad Pablo de Olavide

     Sevilla, España

     
• Localización: Journal of intelligent & robotic systems: Theory & applications, ISSN 0921-0296, Vol. 55, Nº. 4-5, 2009, págs. 323-344
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • This paper presents a new approach for vision-based UAV localization, using mosaics as environment representations. Inter-image motions are used to estimate the motion of the UAV. Online mosaicking is applied to reduce the impact of the accumulative errors in UAV position estimation. A new method to build an stochastic mosaic given the image-to-image homographies is detailed. The mosaic consists of a network of inter-image relations, and is used to create a consistent view of the environment of the UAV and hence, to detect the drift in position estimation by using the mosaic as a resource. The technique could be called simultaneous localization and mosaicking. This technique is specially suitable for monitoring and surveillance tasks in which the UAV will repeatedly cover the same area. The paper also shows experimental results with real UAVs where the benefits of the proposed method are evident.