Citation

Cloudin S, Kumar PM. Int. J. Heavy Veh. Syst. 2018; 25(3/4): 391.

Copyright

(Copyright © 2018, Inderscience Publishers)

DOI

10.1504/IJHVS.2018.094830

PMID

unavailable

Abstract

Vehicular ad-hoc network (VANET) offers a dedicated short-range wireless communication among vehicles on the road and to the road side unit. This paper contributes towards an intelligent decision-making system based on the driver behaviour under normal, reckless, fatigue and drunken conditions, and in developing a mobility model to adapt these support vector machine (SVM) classifier behaviours. Attributes such as speed, accelerometer, alcohol and eye blink values are fed to the mega trend diffusion (MTD) function and the attributes are merged and they are classified using SVM. The basic decisions are made by fuzzy interference system (FIS) which is used to find the driver behaviour and is predicted and appropriate decision will be taken for SVM classification. According to the behaviour of the driver, the speed and direction of the vehicle will be affected and it will be reflected in the mobility pattern of other vehicles. In this work, motion nano sensor was used to sense the values to control the driver behaviour.

Keywords: VANET; driver behaviour; SVM; support vector machine; MTD; mega trend diffusion; FIS; fuzzy interference system; motion nano sensor.

DOI: 10.1504/IJHVS.2018.094830

Language: en