Three Leveled Fuzzy System for Traffic Light and Urban Traffic Control Based on Cellular Automata

Abstract

It’s too years that computational model of cellular automata has been proposed for studying natural phenomena of world; including communication, computation, growing and development, reproducing, contesting and evolution. Vehicular travel which demands on the concurrent operations and parallel activities is increasing throughout the world, particularly in large urban areas. In this paper, to control urban traffic, we study the simulation and optimization of traffic light controllers in a city and present an adaptive fuzzy algorithm based on cellular automata properties. We have used CA for simulating transition function of density of vehicles. In the models that have been proposed till now, environmental factors like priority of intersection streets, width and length of streets and so on, have been assumed equal and therefore they have no role in making decision for changing the status of traffic light, whereas parameters like time during the entire day, density of the vehicles of the street, number of shopping centers, offices, malls,… that have plenty of returnees, have determinant effects on amount of traffic of streets. Considering mentioned notes, we have proposed a novel system that outperforms other available models. Our system has three levels; at the first level, priority of each street is computed momently, based on fuzzy rules and regarding to environmental factors. At the second level, real velocity of vehicles of every street is calculated at specific moment and eventually at the third level, by taking into account two parameters, priority of the street and amount of density behind the traffic light, decision for changing status of traffic light is done. Simulation results of our method have been shown and compared with best algorithms of two most famous available traffic light control approaches -Global and adaptive strategies-.