Abstract

This paper presents a global diffeomorphism from a suitably defined "visible set" of rigid body configurations -- a subset of SE(3) -- to an image-space. Using the diffeomorphism, we develop a purely image-based, essentially global, dynamic visual servoing algorithm that keeps features in the field-of-view, and avoids self-occlusions. The approach is "geometric" in the sense that the visible set and its corresponding image are di#erentiable manifolds, and the diffeomorphism is global. The mapping to image-space relies on a simplified visual target, a sphere marked with a special "arrow" feature point. We present simulations validating application to a more general visual target, a "cloud" of feature points, as is typical in the visual servoing literature.

Citations