Model predictive control of eco-driving for transit using V2I communication

This paper formulates a global eco-driving optimal control of bus transit with considering the platform delay to advise the optimal driving speed and minimizing the volume of pollutant of vehicle emissions, such as NOx, HC, and CO2. The traffic signal timings and waiting passengers at bus stations can be obtained through the communication between vehicles and infrastructure (V2I). Hybrid vehicle dynamic model and dwell time model are proposed to describe trajectories of the bus transit and delay caused by passengers get on and off at each station, respectively. Considering the numbers of passengers at each platform are uncertainty, introducing model predictive control which can updates the current numbers of passengers through V2I at the platform. The global optimal control model is solved with the model predictive control and sub-optimal branch and bound algorithm. An approximately eight-kilometer traffic network including six intersections and seven bus stations of Beijing Bus Line 52 in Chaoyang District, Beijing, is chosen to investigate the performance of the developed eco-driving approach.