Machine Learning Based Real-Time Vehicle Data Analysis for Safe Driving Modeling

This paper identifies a necessity to evaluate the Meta features of vehicles which could be helpful in improving the vehicle driver's skill to prevent accidents and also evaluate the change in the quality of cars over passing time. This paper does an analysis of the vehicle data using supervised learning based linear regression model that is used as an estimator for Driver's Safety Metrics and Economic Driving Metrics. The data collected was obtained from fifteen different drivers over a span of one month which accumulated over 15000 data points. And the metrics that we have devised have potential application in automotive technology analysis for developing an advanced intelligent vehicles. Also, we have presented a system for performing the real-time experiment based on the On-Board-Diagnosis version II (OBD-II) scanner data. Finally, we have analyzed and presented the parameter accuracy over 80% for the driver's safety solution in real-world scenario.