Real-Time Performance Analysis of Multiple Parameters of Automotive Sensor's CAN Data to Predict Vehicle Driving Efficiency

Abstract

Car crashes typically lead to extreme human setbacks and enormous monetary misfortunes in genuine world situations. A convenient exact forecast of car crashes can possibly secure public wellbeing and lessen monetary misfortunes. Driving is a complicated movement whose wellbeing is impacted by a wide scope of variables like driver conduct, vehicle plan and the street climate. Albeit many empowering accomplishments have been made to further develop street wellbeing, yearly 1.35 million individuals bite the dust and upwards of 50 million are harmed and experience long haul inability from street car accidents ("World Health Organization report, 2018,"). The following outskirts of crash anticipation is in the innovation space with an expanding presence of dynamic wellbeing advances like Advanced Driver Assistance Systems (ADAS). A reasonable comprehension of various boundaries affecting driver association with street climate to settle on choices to control their vehicle can give another plan way to deal with a more powerful driver help framework. To investigate the most powerful contributing variables, the proposed work driving style can be analyzed with some safety features and driver-assist features to include, Antilock Braking System (ABS), Traction Control System (TCS), Electronic Stability Control (ESC), Hill Start Assist (HSA) and clutch actuation technique. In the present work, experimental analysis is conducted to evaluate the driver efficiency by using multiple vehicle safety features by acquiring corresponding multiple CAN data wirelessly using Raspberry pi with the ThingSpeak platform. The developed Wireless HMI interface will report vehicle driving patterns and fuel efficiency by giving warning notifications with safety standards to improve the driver's driving style following the road environment.