A GPS-free method for vehicle future movement directions prediction using SOM for VANET

Authors:
Ali Bohlooli;Kamal Jamshidi
Affiliations:
Department of Computer Engineering, Faculty of Engineering, University of Isfahan, Isfahan, Iran 81746-73441;Department of Computer Engineering, Faculty of Engineering, University of Isfahan, Isfahan, Iran 81746-73441
Venue:
Applied Intelligence
Year:
2012

Citing 10
Cited 4

Neural networks: algorithms, applications, and programming techniques

Neural networks: algorithms, applications, and programming techniques
Artificial Neural Networks

Artificial Neural Networks
CarNet: a scalable ad hoc wireless network system

EW 9 Proceedings of the 9th workshop on ACM SIGOPS European workshop: beyond the PC: new challenges for the operating system
Feature Competitive Algorithm for Dimension Reduction of the Self-Organizing Map Input Space

Applied Intelligence
GPS-free Positioning in Mobile Ad Hoc Networks

Cluster Computing
Rough Self Organizing Map

Applied Intelligence
Search on transportation networks for location-based service

Applied Intelligence
Real-time people localization and tracking through fixed stereo vision

Applied Intelligence
A tutorial survey on vehicular ad hoc networks

IEEE Communications Magazine
Clustering of the self-organizing map

IEEE Transactions on Neural Networks

Block-matching algorithm based on harmony search optimization for motion estimation

Applied Intelligence
A Proxy MIPv6 Handover Scheme for Vehicular Ad-hoc Networks

Wireless Personal Communications: An International Journal
A real-time transportation prediction system

Applied Intelligence
A fast prototype reduction method based on template reduction and visualization-induced self-organizing map for nearest neighbor algorithm

Applied Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

Predicting the next movement directions, which will be chosen by the vehicle driver at each junction of a road network, can be used largely in VANET (Vehicular Ad-Hoc Network) applications. The current methods are based on GPS. In a number of VANET applications the GPS service is faced with some obstacles such as high-rise buildings, tunnels, and trees. In this paper, a GPS-free method is proposed to predict the vehicle future movement direction. In this method, vehicle motion paths are described by using the sequence of turning directions on the junctions, and the distances between the junctions. Movement patterns of the vehicles are extracted through clustering of the vehicle's motion paths using SOM (Self Organizing Map). These patterns are then used for predicting the next movement direction, which will be chosen by the driver at the next junction. The obtained results indicate that our GPS-free method is comparable with the GPS-based methods, while having more advantages in different applications regarding urban traffic.