Hierarchical PWARX modeling of obstacle avoidance and steering skill of human driver

The identification of both the states and the skills of drivers has been actively investigated over several decades. The identification of a driver´s skills and its application to driver assistance have been proposed in recent studies. In these studies, it was assumed that different aspects of driving skill can be expressed by switching a number of simple control rules according to the situation. In this study, the driver is assigned the task of steering the vehicle toward a destination while avoiding an obstacle. The steering skill of the driver is evaluated using a previously proposed hierarchical modeling method. The objective of this study is to extract the skill of the driver in a mathematically interpretable form. A number of experiments are conducted in order to verify the validity of the proposed framework.