Abstract
In this report, the problem of driver behaviour modelling is highlighted, followed by the presentation of existing models (Michon, GADGET, DRIVABILITY) and their evaluation. Then, the development of driver behaviour adaptation models and their relation to ADAS are discussed (in terms of automaticity, locus of control, risk homeostasis and compensation, threat avoidance, utility manipulation). The use of driver behaviour models in EC and international projects is analysed, focusing on driver training and assessment, evaluation of safety aids use and ADAS design and impact assessment. The above discussion concludes on the importance of the evaluation and verification of the models output by empirical results. (Author/publisher)