An Unsupervised Approach for Driving Behavior Analysis of Professional Truck Drivers

Modern vehicles can generate up to several Gigabytes of dataper day which are mostly used only for aspects directly related to theproper functioning of the vehicle itself. However, these data have anenormous value as they can be collected and analyzed to better under-stand additional aspects of the driving experience, such as classifying thedriver's behavior and driving style.In this paper, we present a simple yet novel unsupervised methodologythat is able to classify the behavior of a driver in a certain geograph-ical area on the basis of the data collected from all the drivers in thesame area. The proposed methodology has been tested on two differ-ent datasets involving professional truck drivers and it has been verifiedusing human labelled ground truth data. The results obtained demon-strate the feasibility of the proposed solution. To our knowledge, this isthe first study to classify driving behaviours of professional truck driversand validate their performance on such large-scale data with actual safetyscores