Image-based tree detection for autonomous navigation in orchards

  • V. Cadenat Univ. de Toulouse / CNRS / UPS
  • A. Durand-Petiteville UFPE
  • D. Billot Univ. de Toulouse / UPS
  • A. Villemazet CNRS / LAAS

Resumo


This paper deals with the autonomous navigation problem through orchards. It highlights the different challenges arising in this particular environment and proposes an adequate framework. This latter relies on a reactive control strategy fed by visual data provided by a set of RGB-D cameras adequately positioned to enlarge the field of view. In this work, we show interest in coupling a deep learning solution for image-based detection with an existing point cloud processing algorithm to improve the overall perception. Experimental results validate the approach.
Palavras-chave: Agricultural robotics, autonomous navigation
Publicado
09/10/2023
Como Citar

Selecione um Formato
CADENAT, V.; DURAND-PETITEVILLE, A.; BILLOT, D.; VILLEMAZET, A.. Image-based tree detection for autonomous navigation in orchards. In: SIMPÓSIO BRASILEIRO DE ROBÓTICA E SIMPÓSIO LATINO AMERICANO DE ROBÓTICA (SBR/LARS), 15. , 2023, Salvador/BA. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2023 . p. 272-277.