Title
A non-cooperative rear-end collision avoidance scheme for non-connected and heterogeneous environment
Abstract
Rear end collisions are deadliest in nature and result in life losses and severe injuries. Many of the proposed rear-end collision avoidance schemes focus only on specific aspects without considering several non-direct factors, such as driver inattention or distraction, tailgating, panic stops, street surface circumstances, driver response time, person on foot stream and vehicle elements. Henceforth presenting greater challenges in proposing a precise numerical model of the vehicle control framework. Moreover, most collision avoidance schemes incorporate fuzzy logic for formulating human inspired collision avoidance controllers. But the problem with a fuzzy based controller is that their efficiency is greatly affected by the number of rules being deployed. Furthermore, no such agent-based modeling has been utilized, which can effectively model the human-inspired behavior. Literature is full of vehicle control frameworks that work in a connected and homogeneous environment, again critics have raised several questions for such frameworks working in a connected environment. This study aims not only to identify these key issues but also propose a human-inspired physics based Spiral Spring based Simple Reflex Agent for rear end collision avoidance. Our scheme will not only work in the non-connected and heterogeneous environment but will also consider the distraction causing road factors and effectively handle them. Extensive experimentations and presentation for the proof of concept have supported our claims and worthiness of proposed agent-based scheme.
Year
DOI
Venue
2020
10.1016/j.comcom.2019.11.002
Computer Communications
Keywords
Field
DocType
Autonomous vehicles,Agent-based modeling,Collision avoidance,Driver distraction,Spiral spring
Distraction,Control theory,Computer science,Rear-end collision,Fuzzy logic,Response time,Collision,Real-time computing,Proof of concept,Tailgating,Distributed computing
Journal
Volume
ISSN
Citations 
150
0140-3664
1
PageRank 
References 
Authors
0.36
0
5
Name
Order
Citations
PageRank
Sania Khadim110.36
Faisal Riaz2236.03
Sohail Jabbar331434.55
Shehzad Khalid440636.42
moayad aloqaily533137.67