Visual Servoing from Spheres with Paracatadioptric Cameras

Abstract

Summary.A paracatadioptric camera consists of the coupling of a parabolic mirror with a telecentric lens which realizes an orthographic projection to the image sensor. This type of camera provides large field of view images and has therefore potential applications for mobile and aerial robots.This paper is concerned with visual servoing using paracatadioptric cameras. A new optimal combination of visual features is proposed for visual servoing from spheres. Using this combination, a classical control law is proved to be globally stable even in the presence of modeling error. Experimental and simulation results validate the proposed theoretical results.

Improvements on Visual Servoing From Spherical Targets Using a Spherical Projection Model

“…In addition, there is no more a linear relationship between the visual features for translation s t and the camera translational velocities v. Therefore, we prefer the set s n = (s, θu), which is linearly linked w.r.t. v c [see (27)] while enabling a partial control of the target in the image, as discussed in Section III-B.…”

“…It is also possible to show that control law (19) is GAS using s p if and only if R > 0 [27]. Therefore, considering either s or s p , the robustness domain w.r.t.…”

“…the camera can be recovered using the diffeomorphism proposed in [15]. The interaction matrix related to s n has the following form [29]: (27) in which L ω is given by [8] …”

Dynamical Effects of Vision-Based Position Measurement

“…Features of higher order can be also considered. For example, the position of a curved needle was considered as a portion of a circle and located with image processing by Nageotte et al [2005]; visual servoing of spheres is presented by Fomena and Chaumette [2007]. Active vision systems, based on the projection of structured light can also be used for estimation of 3D surfaces as done by Albitar et al [2007].…”

Visual servoing using triangulation with an omnidirectional multi-camera system

“…Another approach to extend the FOV is an omnidirectional vision system, composed of cameras or mirrors, mounted on the robot [7]- [11].…”