Unicycle-like Robots with Eye-in-Hand Monocular Cameras: From PBVS towards IBVS - Visual Servoing via Advanced Numerical Methods - page 334

F AB

F BC = F B ∩ F C

(a)

(b)

Fig. 18.2 (a) topological map: grabbed images are indicated with a capital letter, say I A , I B , I C .
Each grabbed image corresponds to both a certain robot configuration in the metric map (b)
and a node in the topological map. The nodes A and B are connected if and only if the set of
features F AB = F A ∩ F B is not empty; and (b) metric map: the positions A , B , C , representing
each image node in the topological map (a), ξ A , ξ B and ξ C are a set of 3D robot postures

monochromatic camera Jai CVM-50 placed on the robot so that a vertical axis
through the camera pinhole intersects the wheel axis in the midpoint. The camera al-
lows for a

8
was used in the experiment to switch among different controllers. The controller is

Δ = π / 6 semi-aperture of the optical cone, while a threshold of Δ j = π /