Analysis of Car Drivers’ Behaviour and Driving Style

Abstract

Driving security remains one of the important issues. Nowadays, various assistance systems are implemented, such as the systems for analysis of control of a car by its driver. To understand the performance of the driver’s control, a program was created to obtain valuable data and relevant characteristics. To obtain the data, we used an internally designed, laboratory-made vehicle driving simulator developed by D. Michalík [2]. Driver data were obtained using a proprietary vehicle driving simulator, and these were evaluated in the MATLAB environment via integral criteria and other calculated parameters, such as reaction delay. Features thus obtained were used as a training set for the machine learning, using LDA and QDA methods (linear and quadratic discriminant analysis). These methods reveal information concerning the importance of features for the task of driver’s identity prediction based solely on the driving actions.