Modelling topological and metrical properties in physical processes

Abstract

Developing suitable representations for formalising non-trivial domain knowledge has always been central to AI. Within Naive Physics ie. the task of encoding experiential knowledge of the world, few formal theories have appeared that exhibit formal elegance, conciseness and generality to cover a wide variety of modelling problems. We outline a first order formalism being developed that meets these criteria. The formalism is particularly attractive in that it provides the user with the means to model either spatial and/or temporal information as required. The power of the formalism is illustrated by modelling the process of phagocytosis of the amoeba, together with an outline of how many properties of physical entities and relations between them can be modelled within a unitary framework. 1.0 Introduction The importance of representation within a formal framework has always been a central topic for discussion within AI. This has been particularly noticeable since Hayes'...