Robot programs usually let robots interact with objects in their environment. To achieve this, some aspects of those objects need to be described. The basic mechanism for this description builds on the model of Cartesian coordinate frames (also simply referred to as ”‘frames”’). On the one hand, those frames can be used as points defining motions of robots. On the other hand, a frame can be a reference for other frames to build upon. For example, consider some workpiece for which a set of frames is defined, which are used in some series of manipulation tasks. All those frames should be defined relative to some special reference frame of this workpiece. In this case, only this reference frame’s definition needs to be changed if the workpieces location changes. Robot programs that build on some of the other workpiece frames will automatically adapt to their new locations.
The following sections go into details about the representation of frames in the Robotics API, structured as follows: