Appendix F: Object Location and Orientation

The Problem

In this appendix, we present a range of methods that can be used to guide a robot to pick up an object (widget) lying in an unknown position and orientation on a plain-topped table. We shall introduce certain other simplifications:

• 1. The camera is vertically above the widget.

• 2. The widget is entirely within the camera’s field of view and there is no occlusion by any other object.

• 3. If there is more than one object in view, no widget touches any other one.

• 4. There are no artefacts due to poor lighting; there are no significant shadows and there is no glinting from either the widget or table-top.

• 5. There is sufficient contrast to ensure that features that might be used to locate and orientate the widgets are clearly visible.

The layout of the robot vision cell is outlined in Appendix E. We shall assume that the robot has already been calibrated. (Appendix E)

Let us begin with a simple challenge: the child’s toy shown in Fig. F.1 . Despite the fact that a child of about 2...