Identifying Driver Behaviour Through Onboard Diagnostic Using CAN Bus Signals

Abstract

Nowadays, traffic accidents occur due to the increasing number of vehicles. In the researches, it was determined that most of the accidents were caused by the driver. Audible and visual warnings of drivers against possible situations in traffic will reduce the risk of errors and accidents. it was observed that the traffic signs were not enough stimuli for the drivers. For this reason, stimulating electronic applications are developed for drivers in Intelligent Transport Systems. The selection of the correct stimulators by measuring the response of the drivers to different situations in different road conditions will provide a more efficient driving. For this purpose, in order to evaluate the driving behavior of the driver in this study, the speed and RPM information received by means of OBD (Onboard Diagnostic) access to the ECU (Electronic Control Unite) data of the vehicle was evaluated instantaneously. Thus driving information provides aggressive driver detection and warns of traffic hazard situations. For this purpose, an experimental system was created by using machine learning algorithms. The vehicle's speed and RPM data have been used to determine the acceleration of the vehicle and drive. Four different types of drivers have been identified in this designed system. In this way, the driver will be able to detect their own driving. Research will be carried out on how to influence traffic flow by identifying aggressive driver behaviors. It is foreseen that some of the accidents caused by the driver can be prevented.