Driver Behavior Detection Using Smartphone - Dataset

Published: 28 January 2022| Version 2 | DOI: 10.17632/9vr83n7z5j.2
Contributor:
Pawan Wawage

Description

Dataset for driver behavior classification (normal, aggressive, risky) based on accelerometer (X,Y,Z axis in meters per second squared (m/s2)) and gyroscope (X,Y, Z axis in degrees per second (°/s) ) data. Sampling Rate: 50 Hz default Cars: Ford Figo 1.2, Maruti Suzuki Swift VXI, Tata Nexon XMS Drivers: 3 different drivers with the ages between 35-40 yrs. Driver Behaviors: Normal, Aggressive, and Risky Smartphone Sensor: Accelerometer, Gyroscope Smartphone Device: Redmi 4, MI A3

Files

Steps to reproduce

The data was collected on real time by executing the driving events in naturalistic traffic conditions. We used smartphone application named "Sensor Record" to record the sensor data. Raw data is available for every trip. The trip was a round trip and one way trip was approximately 10-25 km.

Institutions

Vishwakarma Institute of Information Technology

Categories

Machine Learning, Driver Behavior, Driver Distraction

Licence