Tutorial: Human factors in modeling a simulation for traffic and transportation: examples in pedestrian and driver modeling

A lot of simulation tools dedicated to the road traffic and the pedestrian crowds were strongly improved during the last decade. At the individual level, the models tend to simulate more accurately the agents (drivers, pedestrians) by increasing the level of details of the simulation, in order to be closer to the real world. On the other side, many works in psychology show that the perception is complex and crucial in order to take the right decision to move in safe way in urban area, whether for drivers or pedestrians. But in the most of models, the perception action is reduced to a very simple and passive action that is to perceive all, or all that is relevant for the agent goals. After a short presentation of the traffic and pedestrian modeling at the microscopic level, this tutorial will address the major role played by the perception in the agent-based simulations for transportation studies. Two examples will be given: the first one deals with the limited perception of drivers in urban intersections, and the second one is focused on the pedestrian crossing behaviour.

Speaker - Jean-Michel Auberlet

Jean-Michel Auberlet conducts his research at IFSTTAR (The French Institute of Science and Technology devoted to Transport, Planning and Networks). He obtained his PhD “Mathematical study of a complex system in physiology: time scales and viability analysis” in 1997. His research fields are related to both perception and interaction modeling in simulation for both the urban traffic and the vulnerable users. He joins The Lepsis laboratory (Laboratory for road Operations, Perception, Simulators and Simulations) of IFSTTAR in November 1999. He co-organized several workshops in the domain of Multi-Agent System and Transportation. He is also a member of the AND30 committee of the Transportation Research Board.

Banner photo by dariuszka via photopin cc