Abstract

Standard vision-based control techniques can be classified into two groups:model-based and model-free visual servoing.Model-based visual servoing is used whcn a 3D model of the olserved object is availablc.If thc 3D modcl is complctely unknown,robot positioning can still be achievedl using a teaching-by-showing approach.This mnodel-frcc tcchniquc nccds a prcliminary lcarning stcp during which a rcfcrcncc imnage of the scene is stored.The objective of this paper is to propose an unificd approach to vision-bascd control which can bc uscd with a zoomning camnera whether the model of the object is known or not.The key idea of the unified approach is to build a reference in a projective space invariant to camera intrinsic parameters which can be computed if the model is known or if an image of the objcct is availablc.Thus,only one low level visual servoing techique inust be imnpleinented at ouce