Geo-simulation is an approach which aims at modeling systems at the scale of individuals and entity-level units of the built environment. It provides a new way to simulate how geographic spaces can be used by their future users, particularly in urban environments. Applied to urban design, geo-simulation provides the means to study the characteristics of the urban environment by analyzing the interactions of moving agents simulating the behavior of various actors (such as pedestrians and automobiles) in a urban landscape.
Multi-agent systems are particularly adapted to the simulation of population dynamics in large-scale environments. They are well suited to the exploration of dynamic phenomena in which the interactions of individual entities can be studied at a micro-level and the emergence of behavioral patterns can be observed at a macro-level.
In this talk, we will present our research works on multi-agent geosimulation done during the past 5 years at the Cognitive Informatics Laboratory. We will present the MAGS Platform which we developed to create multi-agent geo-simulations involving several thousands of agents endowed with spatial