Agents Cognitifs Territorialisés pour l’Etude des dynamiques Urbaines et des Risques (ACTEUR)

The use of agent-based models to study urban systems is booming for the last ten years. Some recent models propose to include cognitive agents, i.e. agents able to make complex reasoning, to improve the simulation realism. However, developing such models require high-level programming skills. Indeed, if there are nowadays numerous software platforms that ease the development of models, none of them are adapted to the development of large-scale models that integrate cognitive agents by non-computer scientists. The objective of this project is to develop a platform to help modelers to design cognitive agents through a graphical language. The platform has also for ambition to be used as a support of model discussion between stakeholders. The platform will be used for two case studies concerning La Réunion island urban evolution and the adaption to industrial hazards in Rouen.

Objectives

Carry out innovative urban planning policies to face urban residents growth using a generic modelling and simulation platform.

Urban Growth

Final Product

  • A simulation and modeling open­source platform dedicated to the design of realist models integrating territorial cognitive agents.

  • A user guide for the platform

  • A model concerning the simulation of the impacts of industrial hazards on the urban mobility for the city of Rouen

  • A model concerning the simulation of urban evolution applied for La Réunion Island

Contributions

Agent-Based Simulations

  • A new cognitive agent architecture enabling to simulate hundreds of thousands agents

  • A graphical language to design cognitive agents

  • Methods to analyze and calibrate cognitive agents’ behaviors

Geographical sciences

  • A complete modeling methodology for building urban models

  • Recommendations for emergency management in case of industrial hazards in the city of Rouen

  • Recommendations for new urban planning policies in La Reunion Island