Dark Mode
Road Behaviour Classification
Data Science and Analytics
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This dataset is designed for predicting driving behaviour, specifically focusing on aggressive driving. Aggressive driving is identified as a primary cause of road traffic accidents, accounting for over 55% of fatal crashes during a recent four-year period. The dataset offers a solution to accurately and quickly predict such dangerous driving actions, including speeding, sudden braking, and sharp turns. The data is collected via accelerometer and gyroscope sensors from smartphones, leveraging their ubiquity.
Columns
- AccX: Acceleration along the X-axis, measured in metres per second squared (m/s²).
- AccY: Acceleration along the Y-axis, measured in metres per second squared (m/s²).
- AccZ: Acceleration along the Z-axis, measured in metres per second squared (m/s²).
- GyroX: Rotation along the X-axis, measured in degrees per second (°/s).
- GyroY: Rotation along the Y-axis, measured in degrees per second (°/s).
- GyroZ: Rotation along the Z-axis, measured in degrees per second (°/s).
- Class: A classification label categorising driving behaviour as 'SLOW', 'NORMAL', or 'AGGRESSIVE'.
- Timestamp: The time in seconds, representing the unit of time provided by the Android sensors.
Distribution
The dataset is typically provided in a CSV file format, with a sample file named
test_motion_data.csv available, which is 257.92 kB in size. It comprises 8 columns and contains 3084 records. The data was collected at a sampling rate of 2 samples (rows) per second. Gravitational acceleration has been removed from the sensor readings. The data was collected using a Samsung Galaxy S21 device.Usage
This dataset is ideal for developing predictive models to identify and classify dangerous driving behaviour. It can be applied in machine learning and deep learning projects aimed at enhancing road safety, driver monitoring systems, and accident prevention. Potential uses include building driving behaviour datasets and conducting driver behaviour analysis.
Coverage
The dataset focuses on capturing various aspects of driving behaviour through accelerometer and gyroscope sensor data. While specific geographic or demographic scopes are not detailed, the data collection method using common smartphone sensors suggests broad applicability. The data reflects driving actions such as speeding, sudden braking, and sudden turns.
License
Attribution 4.0 International (CC BY 4.0)
Who Can Use It
This dataset is suitable for data scientists, machine learning engineers, and researchers interested in automotive safety, behavioural analytics, and sensor data processing. It can be used by those looking to develop algorithms for predicting aggressive driving, conduct studies on driver behaviour, or integrate intelligent safety features into applications.
Dataset Name Suggestions
- Aggressive Driving Sensor Data
- Smartphone Driving Behaviour
- Vehicle Motion Prediction
- Driver Safety Analytics
- Road Behaviour Classification
Attributes
Original Data Source: Road Behaviour Classification
QUALITY