A hybrid visual servoing control of 4 DOFs space robot

Abstract

This paper presents research and development for a hybrid visual servoing control method for a newly developed space robot. The hybrid scheme consists of an eye-in-hand camera and eye-to-hand camera configuration. The first one guarantees good accuracy and the ability to explore the workspace; the second one ensures a panoramic sight of the workspace. The motivation of this paper is to take advantage of both in a cooperation scheme to makes the execution of complex tasks easier and offers higher flexibility in the presence of a dynamic scenario. Compared with other multi-camera system, both global and local vision has the capability of determining pose of target independently. Moreover, this paper also addresses suitable object recognition and pose estimation algorithm based on polygonal approximation. To realize real-time visual servoing control, part of recognition module has been optimized to increase process speed. Finally, 3D visual servoing within a dynamic look and move scheme based on estimated object pose has been carried out. Experimental results demonstrate the feasibility and the effectiveness of the proposed method.