Critical reasoning: AI for emergency response

摘要

Effective response to emergencies depends upon the availability of accurate and focused information. The goal of the FireGrid project is to provide an architecture by which the results of computer models of physical phenomena can be made available to decision-makers leading the response to fire emergencies in the built environment. In this paper we discuss the application of a number of AI techniques in the development of FireGrid systems, and include algorithms developed for reasoning about dynamic situations. It is intended that this paper will be of technical interest to those who have to construct agents that are able to reason about the complexities of the real world, and of more general appeal to those interested in the ontological and representational commitments and compromises that underlie this reasoning.