A simulation based interaction analysis of automated vehicles

Autonomous vehicles will increasingly shape the streetscape in the future. At the beginning of the dissemination, interactions between autonomous vehicles will occur without a predefined communication interface. In this work, a simulation framework is set up that allows the investigation of the interaction of autonomous vehicles. In particular, an EGO centric vehicle simulation is converted into a vehicle simulation with multiple detailed modeled vehicles (MultiEGO) using co-simulation methods. To identify the risk in an early stage by using simulation methods, a criticality assessment of traffic situations is defined. In a case study, critical traffic situations are analyzed using a detailed vehicle model and a search method based on a generalized traffic scenario. This investigation shows how demanding it is to find all the driving malfunctions and gives an insight into the challenges of developing and releasing autonomous vehicles.