Foundations of the Socio-physical Model of Activities (SOMA) for\n Autonomous Robotic Agents

Abstract

In this paper, we present foundations of the Socio-physical Model of\nActivities (SOMA). SOMA represents both the physical as well as the social\ncontext of everyday activities. Such tasks seem to be trivial for humans,\nhowever, they pose severe problems for artificial agents. For starters, a\nnatural language command requesting something will leave many pieces of\ninformation necessary for performing the task unspecified. Humans can solve\nsuch problems fast as we reduce the search space by recourse to prior knowledge\nsuch as a connected collection of plans that describe how certain goals can be\nachieved at various levels of abstraction. Rather than enumerating fine-grained\nphysical contexts SOMA sets out to include socially constructed knowledge about\nthe functions of actions to achieve a variety of goals or the roles objects can\nplay in a given situation. As the human cognition system is capable of\ngeneralizing experiences into abstract knowledge pieces applicable to novel\nsituations, we argue that both physical and social context need be modeled to\ntackle these challenges in a general manner. This is represented by the link\nbetween the physical and social context in SOMA where relationships are\nestablished between occurrences and generalizations of them, which has been\ndemonstrated in several use cases that validate SOMA.\n