The Concept of Determining Route Signatures in Urban and Extra-Urban Driving Conditions Using Artificial Intelligence Methods

Abstract

The article describes the implementation of road driving tests with a vehicle in urban and extra-urban traffic conditions. Descriptions of the hardware and software needed for archiving the data obtained from the vehicle’s on-board diagnostic connector are presented. Then, the routes are analyzed using artificial intelligence methods. In this article, the reference of the route was defined as the trajectory of the driving process, represented by the engine rotational speed, the driving speed, and acceleration in the state space. The state space was separated into classes based on the results of the cluster analysis. In the experiment, five classes were clustered. The K-Means clustering algorithm was employed to determine the clusters in the variant without prior labelling of the classes using the teaching method and without participation of a teacher. In this way, the trajectories of the driving process in the five-state state space were determined. The article compares the signatures of routes created in urban and extra-urban driving conditions. Significant differences between the obtained results were indicated. Interesting methods of displaying the saved data are presented and the potential practical applications of the proposed method are indicated.