Dissertation Talk: Dan Calderone: Models of Competition for Smart Cities: Routing Games, Mean-field Games, and Varying Population Preferences

Lecture | February 24 | 9-10 a.m. | Cory Hall, 521 Hogan Room

 Dan Calderone

 Electrical Engineering and Computer Sciences (EECS)

Abstract:
Competition underlies much of the complexity of modern transportation systems and correctly modeling the incentives that transportation users face is critical in designing the smart cities of tomorrow. In the past century, classical routing games have proved a powerful tool in modeling competition in transportation. In this talk, we explore several extensions of routing games to model scenarios relevant to smart cities. Particularly, we explore connections between routing games and mean-field games and develop a framework for analyzing how varying preferences among a population affect the traffic equilibrium. We use these extensions to model city drivers looking for on-street parking, ride-sharing drivers competing for customers, and commuters weighing external factors (such as money, convenience, and privacy) along with travel time in their routing decisions.

 CA, danjc@eecs.berkeley.edu, 2406207849