FPGA urban traffic control simulation and evaluation platform

Abstract

The study and development towards Urban Traffic Management and Control (UTMC) Systems have not solely or recently gained extreme importance only due to obvious issues such as traffic safety improvement, traffic congestion control and avoidance but also due to other underlying factors such as urban transportation efficiency, urban traffic originated air pollution and future concepts as are autonomous vehicle systems, which are presently taking shape. Generally speaking urban traffic simulations occur in a software environment, which comes to hinder the progress taken towards the actual implementation of UTMC systems. The reason to why such happens is based on the fact that urban traffic controllers are usually implemented and executed on hardware platforms, therefore software based models don‟t support an actual implementation directly. In this study we explore a novel approach to urban traffic simulation, aimed to eliminate the timeframe and work-distance between the UTMC system‟s design and an eventual implementation, where a Field Programmable Gate Array (FPGA) is used to execute a simulation model of an urban traffic network. Since the resource to FPGAs implies a hardware based execution, the resulting implementation of each traffic management and control element can be considered not only as having a close matched behavior to a real world implementation but also as an actual prototype. From the simulation viewpoint the use of FPGA‟s holds the prospect of being able to hold execution speeds many times faster than software based simulations as FPGA designs are able to execute a large number of parallel processes. This study shows that an Urban Traffic Control Simulation and Test Platform is possible by implementing a relatively simple urban network model in a low end FPGA. This result implies that with further time and resource investments a rather complex system can be developed which can handle large scale and complex UTMC systems with the promise of shortening the work distance between the concept and a real world running implementation.