Daisy Chaining Based Visual Servo Control Part II: Extensions, Applications and Open Problems

In this paper, the open problems and applications of a daisy chaining visual servo control strategy are given. This paper is Part II of Hu et al. (2007) in which a tracking problem using the daisy chaining strategy is addressed. The main idea of the daisy chaining strategy is to use multi-view geometry to relate coordinate frames attached to the moving camera, moving planar patch, and the desired planar patch specified by an a priori image. Geometric constructs developed for traditional camera-in-hand problems are fused with fixed-camera geometry to develop a set of Euclidean homographies. Based on the homographies, the corresponding rotation and translation components can be extracted for use in the control development. Different from the traditional camera-to-hand and camera-in-hand visual servo control configurations, two cameras are used to construct the homography relationships and estimate the pose of an object modeled as a planar patch (e.g., an unmanned ground vehicle (UGV) or an unmanned air vehicle (UAV)) when in case the object is out of the field of view (FOV), or when the current and desired poses of the object are within the FOV of a single camera.