Driving pattern analysis to determine driver behaviours for a local authority based on the cloud using OBD II

Abstract

Aggressive driving is the leading cause of road accidents, which mainly results from driving behaviour that endangers drivers and the people around them. Such driver behaviours should be identified by local authorities to correct the behaviours or understanding the root cause ofaccidents. Data recordedusing the On BoardDiagnostic (OBD)IIdevice can be analyzedto determine such behaviours. A sudden change in the manoeuvring of a vehicle indicates aggressive driving behaviour, which eventually yields non-uniform parameter values returned by the engine control unit (ECU) system without any specific reason. In this study, real-time data were recorded from ECU by using OBD II and an accelerometer. Artificial Intelligence was used to group different types of data to identify behaviours based on the similarity of data points. This study aimed to identify such drivers and reduce the risk of accidents. Driving behaviours were categorized as bad, normal, and aggressive. Because clustering is based on crowded data, which signifies similar driving patterns most of the time in the course of recording, this study used density-based spatial clustering of applications with noise unsupervised learning algorithm. The system sends data to the cloud, enabling the authorities to access it from any place for further action. ANOVA was conducted using IBM's Statistical Package for Social Sciences to compare and determine the most favourable method to collect data.