Characterizing the "driver DNA" through CAN bus data analysis

People's driving behavior is influenced by different human and environmental factors, and several attempts to characterize it have been proposed. Nowadays, the standardization of the CAN bus and the increase of the electronic components units in modern cars offer a large availability of sensors data that make possible a more reliable and direct characterization of driving styles. In this work, we propose the concept of "Driving DNA" as a way of describing the complexity of driving behavior through a set of individual and easy-to-measure quantities. These quantities are responsible for some aspects of the driver's behavior, just as - in the metaphor - genes are responsible for the tracts of an individual. The concept has been tested on a dataset collected from the CAN bus consisting of more than 2000 trips performed by 53 people, in a wide scenario of road types and open traffic conditions. The Driving DNAs have been calculated for each person, and a graphical visualization of their comparison is provided.