Year: 2025
Type: conference-paper
Abstract: This paper presents a visual servoing strategy for a clevis and tenon joint assembly as typically performed in aircraft manufacturing. The desired velocity in the image acquired by the cameras observing the parts is extracted from a vector field that guides the motion of the moving part so that it follows an accurate 3D straight line trajectory. The vector field is designed so that it is continuous and globally exponentially stable whatever the initial configuration. The strategy has been successfully implemented and validated on a full-scale demonstrator, with coarse camera localization, while ensuring an assembly without any collision thanks to a maximal error less than 1 mm along a 10 cm trajectory.
Institutions: IRT Jules Verne, Nantes Université, Centre National de la Recherche Scientifique, Institut national de recherche en sciences et technologies du numérique, Institut de Recherche en Informatique et Systèmes Aléatoires, RAINBOW: robotique interactive et référencée capteurs, Université de Rennes
Language: English
Topic: Advanced Vision and Imaging
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4414384049
FWCI:
Cites:
Cited by:
Related to: