Lane change behaviour recognition using neural entwork / N. J. Zakaria ... [et al.]

Zakaria, N. J. and Zamzuri, H. and Mohamed Ariff, M. H. and Azmi, M. Z and Hassan, N. (2018) Lane change behaviour recognition using neural entwork / N. J. Zakaria ... [et al.]. Journal of Mechanical Engineering (JMechE), SI 6 (1). pp. 1-19. ISSN 18235514

Abstract

Lane change behaviour recognition is one of the significant elements in advanced vehicle active system for the purpose of collision avoidance and traffic flow stability to ensure a safer driving experience. The system recognizes either the driver in situations of normal or evasive lane change maneuver which respond and assist the driver negligence. This paper proposes a lane change behaviour recognition using Artificial Neural Network (ANN) model by classifying the behaviour either evasive or normal lane change. An ANN model was adopted in order to combine several vehicle state information to generate the lane change behaviour classification. The vehicle state parameters such as vehicle speed, yaw rate, time taken for one complete steer cycle and steering angle were used as the inputs to develop in the ANN model. The state parameters were acquired from a real-time experiment conducted by several selected normal drivers. The result shows that the proposed ANN model has successfully recognized 94% and 92.8% of the lane change samples in training and test data set respectively. Hence, the proposed ANN model has a promising potential to handle system nonlinearity.

Metadata

Item Type: Article
Creators:
Creators
Email / ID Num.
Zakaria, N. J.
UNSPECIFIED
Zamzuri, H.
UNSPECIFIED
Mohamed Ariff, M. H.
UNSPECIFIED
Azmi, M. Z
UNSPECIFIED
Hassan, N.
UNSPECIFIED
Subjects: T Technology > TA Engineering. Civil engineering > Engineering mathematics. Engineering analysis
T Technology > TJ Mechanical engineering and machinery
Divisions: Universiti Teknologi MARA, Shah Alam > Faculty of Mechanical Engineering
Journal or Publication Title: Journal of Mechanical Engineering (JMechE)
UiTM Journal Collections: UiTM Journal > Journal of Mechanical Engineering (JMechE)
ISSN: 18235514
Volume: SI 6
Number: 1
Page Range: pp. 1-19
Keywords: Lane Change, Artificial Neural Network (ANN) Model, Normal, Evasive.
Date: 2018
URI: https://ir.uitm.edu.my/id/eprint/40990
Edit Item
Edit Item

Download

[thumbnail of 40990.pdf] Text
40990.pdf

Download (984kB)

ID Number

40990

Indexing

|

Statistic

Statistic details