Multi-Agent System Simulation of Indoor Scenarios

Abstract

This paper presents a flexible agent decision model for the simulation of indoor scenarios. There are different kind of applications with varying requirements, from the typical emergency evacuation where the physical interaction of crowds of agents are more relevant, to those that demand more sophisticated agent decision models as when testing smart environment applications. Existing tools usually focus on one of these issues, looking for efficiency in the solutions. The agent decision model in this paper tries to get a balance between efficiency and flexibility in the specification of the agent behavior in simple and complex situations. This is applied in a simulation framework for indoor scenarios, although it could be extended to other settings.