Date

2016

Author

Kumtepe, Ömürcan

Metadata

Show full item record

Item Usage Stats

148
views

81
downloads

Cite This

Every year a vast number of traffic accidents occur globally. These traffic accidents cause fatalities, severe injuries and huge economical cost. Most of these traffic accidents occur due to aggressive driving behaviour. Therefore, detection of driver aggressiveness could help reducing the number of traffic accidents by warning related authorities to take necessary precautions. Although aggressiveness is a psychological phenomenon, driver aggressiveness can be analysed by monitoring certain driving behaviour such as abrupt lane changes, unsafe following distance and excess acceleration and deceleration. In this thesis work, a method is introduced in order to detect aggressive driving behaviour using a system on vehicle. The proposed method is based on fusion of visual and other sensor information to characterize related driving session and to decide whether the session involves aggressive driving behaviour. Visual information is used to detect road lines and vehicle images; whereas CAN bus information provides certain driving data such as vehicle speed and engine speed. Both information is used to obtain feature vectors which represent a driving session. These feature vectors are obtained by modelling time series data by Gaussian distributions. An SVM classifier is utilized to classify the feature vectors in order for aggressiveness decision. The proposed system is tested by real traffic data and it achieved an aggressive driving detection rate of 94.0%.

Subject Keywords

Automobile driving., Motor vehicle driving., Traffic engineering., Global Positioning System., Neural networks (Computer science).

URI

http://etd.lib.metu.edu.tr/upload/12619712/index.pdf
https://hdl.handle.net/11511/25403

Collections

Graduate School of Natural and Applied Sciences, Thesis