Simulation-Based Analysis of a Network Model for Autonomous Vehicles with Vehicle-to-Vehicle Communication

Abstract

Autonomous vehicle technology is an expansively researched area of transport that aims to tackle long-standing problems of traffic such as congestion, safety and efficiency. Many of these vehicles combine automated driving with communication among vehicles and infrastructure to bring about a seamless driving experience that would not have been possible with human driving. The development of computer simulations for such vehicles aims to address concerns on whether the benefits proposed by autonomous vehicle makers can be realized in various traffic environments. This paper assesses the efficiency of autonomous vehicles that are introduced on an arterial road network with features similar to Singapore’s road networks. A cellular automata simulation has been developed that considers vehicle-to-vehicle communication abilities of autonomous vehicles. A traffic data collection algorithm based on web traffic services was developed to estimate real-time travel times along each stretch of road in the network simulation, from which autonomous vehicles can optimize their speed and route for a faster journey time. Based on preliminary results, the simulation was tested under multiple traffic densities and situations. The results display interesting interactions between vehicles and road elements such as lanes and traffic lights, which has allowed both autonomous and non-autonomous vehicles to travel to their designated destination faster when autonomous vehicles have been introduced.