Towards Affordance-Based Robot Control

Abstract

Gibson's notion of affordance seems to attract roboticists' attention. On a phenomenological level, it allows functions, which have "somehow" been implemented, to be described using a new terminology. However, that does not mean that the affordance notion is of help for building robots and their controllers. This paper explores viewing an af-fordance as an abstraction from a robot-environment relation that is of inter-individual use, but requires an individual implementation. Therefore , the notion of affordance helps share environment representations and theories among robots. Examples are given for navigability, as afforded by environments of different types to robots of different undercarriages and sensor configurations. 1 Background Among the most basic properties that a mobile robot needs to perceive about its environment is whether it can go someplace, i.e., drive, walk, crawl, climb-whatever its kinematics. State-of-the-art work in indoor robot navigation typically abstracts away from many details of the problem by assuming implicitly that any area that is not perceived as being blocked is navigable. That works as long as you make sure your robot stays clear of staircases, glass doors and mirrors, no drawers out of sensor height are left open, the robot fits under all table-tops, and, of course, the floor is sufficiently flat for its undercarriage. In a nutshell: Out of relatively protected lab environments, it is today non-trivial for a mobile robot to determine whether it can physically move to some location in its vicinity.