Visual Servoing from Spheres using a Spherical Projection Model

Abstract

In this paper, we investigate the use of a spherical projection model to search for optimal visual features for visual servoing. A new minimal set of three visual features is proposed for visual servoing from spheres using any central catadioptric system. Using this set of features, a classical control method is proved to be globally stable even in the presence of modeling error and locally stable to calibration errors on perspective cameras. Using this type of cameras, experimental results are presented and validate the proposed theoretical results.