On Discovering Road Traffic Information Using Virtual Reality Simulations

Transport systems play critical role in modern day economies and are prime example of highly dynamic complex systems. Current traffic monitoring systems do not provide timely and granular information which could sufficiently meet system analysis requirements. Simulation augmented with virtual reality (VR) has emerged as an important methodology to model and analyze transport systems. This paper-reports our ongoing work on developing an open source road traffic VR simulator based on microscopic and car-following models, designed using Java Monkey Engine. Subsequent to a brief literature review on traffic simulations, we give a description of the simulator design, its functionality, and the driver behaviour models used in the simulator. We then simulate few interesting traffic scenarios and analyse traffic flow,speed, occupancy and other results produced by the simulator. Finally, we conclude and outline the future work.

I. Introduction

Modem societies all around the world have developed transportation systems to enable its citizens to move around freely and to help economies grow [1], [2]. Transport systems play critical role in modern day economies and are a prime example of highly dynamic complex systems. With rapidly increasing traffic demands, congestion and road safety have become a major concern causing each year human and economic loss. The estimated number of deaths caused by traffic accidents in the US alone is over 42 thousands per year (US Traffic Safety Annual Assessment 2006) adding up to the 1.2 million deaths annually in the world (World Health Organisation, 2004). Road congestion causes UK economy an estimated loss of £20 billion annually (Confederation of British Industry). Various traffic monitoring systems such as those based on induction loops and Automatic Number Plate Recognition (ANPR) systems have been in place to monitor, control and optimise road traffic. However, these systems do not provide timely and granular information sufficient to meet demands for the next generation transportation systems. On the other hand, the sheer amount of data [3] generated by the current transportation systems has been a challenge due to lack of realistic traffic models and huge computational resources that data analysis requires [4]. Simulation has emerged as an important methodology to model and analyze existing and future transport systems. Simulation when augmented with Virtual Reality (VR) and visualisation provides enhanced ability for the users to immerse and interact with the simulated system, and to analyse complex systems with ease and in greater details.