Mean field differential games in intelligent transportation systems

In this paper, we study Intelligent Transportation Systems (ITS) for future Smart Cities. We propose a new type of Automated Highway System (AHS), which optimizes both total consumed energy and travel time. We build on our previous work by adapting it to ITS as a new class of games, which is Mean Field (MF) Multilayer Stackelberg Games (MSG). In this game model, a leader (Control Center) imposes its fixed strategy on all followers (Vehicles). However, the Control Center cannot manipulate the vehicles directly. Instead, sub-leaders (Road Links) link up vehicles to the control center as an intermediary layer. Therefore, vehicles of each road link are weakly coupled through a Mean Field term, which affects the vehicles' individual costs. The main novelty in this paper is to consider the Mean Field Game for Automated Highway Systems in a Smart City. Because of the fact that there are no fully autonomous highways in cities, implementing new type the algorithm on ITS is nearly impossible for real life scenarios. Hence we have developed our highly realistic simulation environment, Istanbul Technical University City Transportation Simulator (ITUCiTSim) program. We demonstrate the effectiveness of the proposed method on ITUCiTSim.