Modeling of Driver's Longitudinal Behavior

In the last years, many vehicle manufacturers have introduced advancedriver support in some of their automobiles. One of those new featuresis Adaptive Cruise Control (ACC), which extends the conventionalcruise control system to control of relative speed and distance toother vehicles. In order to design an ACC controller it is suitable tohave a model on drivers' behavior. Our approach to find dynamicalmodels of the drivers' behavior was to use system identification.Basic data analysis was made by means of system identificationmethodology, and several models of drivers' longitudinal behavior areproposed, including both linear regression models and subspace basedmodels. Detection when a driver is changing his behavior in varioussituations to a deviant behavior is useful. To that purpose a GARCHmodel was used to model the driver in arousal situations, where thedriver changes behavior, is proposed.