Driver's steering behaviour identification and modelling in near rear-end collision

• Main Authors: Azmi, M.Z (Author), Hassan, N. (Author), Wahid, N. (Author), Zamzuri, H. (Author), Zulkepli, K.A (Author)
• Format: Article
• Language: English
• Published: Universitas Ahmad Dahlan 2017
• Subjects: Artifitial neural network; Driver behaviour modelling; Driver steering manoeuvre; Rear-end collision avoidance
• Online Access: View Fulltext in Publisher; View in Scopus
• ISBN: 16936930 (ISSN)
• ISSN: 16936930 (ISSN)
• DOI: 10.12928/TELKOMNIKA.v15i2.6133

This paper studies and identifies driver's steering manoeuvre behaviour in near rear-end collision. Time-To-Collision (TTC) is utilized in defining driver's emergency threat assessment. The target scenario is set up under real experimental environment and the naturalistic data from the experiment are collected. Four normal drivers are employed for the experiment to perform the manoeuvre. Artificial Neural Network (ANN) is proposed to model the behaviour of the driver`s steering manoeuvre. The results show that all drivers manage to perform steering manoeuvre within the safe TTC region and the modelling results from ANN are reasonably positive. With further studies and improvements, this model would benefit to evaluate the driving reliability to enhance traffic safety and Intelligent Transportation System. © 2017 Universitas Ahmad Dahlan.