Affordances are commonly described as opportunities for action; features of an agent's environment that guide the agent's behavior. The concept of affordance is employed in a range of fields, including ecological psychology, artificial intelligence, and human factors. Yet there is no widely accepted, concrete formalization of the concept. Our goal is the development of a practical and theoretically well-motivated account of affordances, appropriate for embodied habile (i.e., tool-using) agents.;In a typical tool use task, we can view both the relationship between the agent and the tool and the relationship between the tool and the target in terms of affordances. One set of affordances relates to the ability of the agent to manipulate the tool, while a second set of affordances relates to the ability of the agent to manipulate the target by means of the tool. In both cases, effective tool use is facilitated by the coupling of one object to another: agent-to-tool-to-target. Thus we will refer to the subset of affordances that support such interactions as "coupling" affordances.;In the work presented here, we focus on the visual identification of such affordances via contour similarity. Objects with complementary contour segments can fit together, which suggests possible opportunities for effective interactions. We present a system for the identification and evaluation of partial contour-based matches and analyze the system's behavior. Noting the lack of an established framework for the evaluation of such a system, we propose a set of sample tool-use scenarios as part of our analysis. Finally, we demonstrate the use of the system in providing guidance to an autonomous robotic agent performing tool selection tasks.
展开▼
