A Large-Scale Traffic Scenario of Berlin for Evaluating Smart Mobility Applications

Research on novel concepts in the field of smart mobility and ITS requires employing traffic simulations in combination with communication and application simulations. With Eclipse MOSAIC we developed a co-simulation simulation framework to setup holistic system simulations in that very field, by coupling best-in-class simulators from various research domains. One important task here is modeling road traffic, which is non-trivial on a large scale. Traffic for a city-wide area can be modeled on a macroscopic or microscopic level, however, only the later provides realistic vehicle movements which is a requirement for communication and application simulation. Currently, there are only a handful of scenarios that model enough traffic to reliably test smart mobility applications. In this paper, we describe how we created a large-scale simulation scenario depicting a full day of motorized private traffic for the City of Berlin. To achieve this, we created a scenario for the microscopic traffic simulator SUMO, with the traffic demand extracted from an existing MATSim scenario and transferred to SUMO using iterative traffic assignment. Comparing the simulated counts with real data emphasizes that this scenario can model traffic in Berlin close to reality. With more than 2.2 million trips within an area of 800 km2 this is the largest traffic scenario we are currently aware of, and we will provide it for other researchers under an open-source license.