Implementation of cognitive driver models in microscopic traffic simulations

In order to perform microscopic traffic simulations as realistically as possible, a detailed modelling of each individual driver is essential. Currently established microscopic traffic simulators follow a rather static approach to model the driver's behaviour. Individual emotional influences and temporarily occurring distracting factors are heavy to implement in the established state-of-the-art microscopic traffic simulators. This paper proposes a solution how emotional influences and distracting factors can be integrated in established traffic simulation tools. For this purpose, robot-learning approaches are adapted to model the emotional state of vehicle drivers. In the end of this work, a proof of concept is done to illustrate the strength of the developed approach.