THE INTERNET OF CARS
ABSTRACT: The “Internet of Cars” refers to a cyber-physical system consisting of Connected Automated Vehicles (CAVs) whose ultimate goal includes automating all aspects of mobility, from interconnected self-driving vehicles to on-demand sharing of transportation resources. The availability of large amounts of data, ubiquitous wireless connectivity, and the critical need for scalability open the door for new control and optimization methods that will enable the creation and effective operation of such a system. The talk will address two key questions. First, how can we quantify the expected benefits of an Internet of Cars so as to justify the challenging technological, economic, and social transitions involved towards its implementation? To do so, we will describe how large amounts of actual traffic data can be harnessed and drive inverse optimization methods to measure the value of CAVs in terms of eliminating the prevailing Price of Anarchy: the gap between current “selfish” user-centric traffic equilibria and optimal “social” system-centric equilibria which are achievable with automated mobility. Second, a decentralized optimal control framework will be presented to show how CAVs can operate to achieve the goals of reducing congestion and energy consumption while ensuring passenger comfort and guaranteeing safety requirements. As an illustration, we will show how to accomplish this at signal-free urban intersections.