leimazizizi / pedsimcity Goto Github PK

This model incoporates simplified cognitive maps of the urban environment in the architecture of the agents In particular, a computational approach to Kevyn Lynch's The Image of the City (see related paper in Cities) is employed to incorporate salient urban element in the route-choice approaches of the agents.

The model is built on Mason and GeoMason 1.6.

A basic conceptual model of the ABM was presented at the 2018 Agile Conference.

The introduction of different urban elements has been tested, in combination with existing route choice models:

• Landmark-based navigation: London - landmarkBased branch - Methods, results and evaluation are documented in Modelling the effect of landmarks on pedestrian dynamics (in publication in Computers, Environment and Urban Systems).
• Region- and barrier-based navigation: London and Paris - regionBased branch - Methods and results, along with a validation are documented in Perception of urban subdivisions in pedestrian movement simulation (in publication in PLoS ONE).

It requires:

• GeoMason and its requirements;
• UrbanMason, a set of supporting functions for urban simulation models.

Before running the model, check the default parameters in the class UserParameters.java so to run the ABM as you please; there, one can also set the local directories for storing the output.

The core of the simulation (the simulation state, the scheduler and the agent basic movement functions) is based on Gridlock, by Sarah Wise, Mark Coletti, and Andrew Crooks, a GeoMason example.