A Fuzzy Logic-Based Smart Traffic Management Systems

Abstract

Traffic congestion is a serious problem in many developing countries like Egypt. In the presence of the traditional traffic light system, it creates health risks, wastes time and fuel, and pollutes the air. Therefore, there is a big need for a smart traffic management system. This study contributes in solving this problem by introducing an artificial intelligence-based smart traffic light system using fuzzy logic to ensure the smooth flow of traffic in cities. Research results indicate that smart traffic light management is very crucial in smart cities to reduce their carbon footprint to save the world and save energy. Other possible application could be implemented with minor system upgrades such as Detection and Management of traffic Congestion, Automatic Billing of Toll Charges, Automatic detection of speed limit Violation, Route planning, Intelligent Internet of Vehicles, and Prevention of Road Accidents. Four traffic light approaches (fixed-time, smart-time, fixedtime-fuzzy logic, and smart-time-fuzzy-logic) are developed with different eight traffic scenarios. The four approaches are applied with two tools, the first tool is a real Marquette which consists of 4 roads with only one or two vehicles. The second is a simulation software which consists of 4 roads with more than 100 vehicles through 10 min. Many sensors such as (IR sensors, rain sensors, and LEDs to control traffic light) are used to collect data and then the fuzzy algorithm is used to ensure the smooth flow of traffic in cities. The traffic data acquired from the vehicles is fed into the proposed model to maximize the duration of the green light based on the road state. According to experimental results, the proposed smart fuzzy technique simulation software reduced the average waiting time of the vehicles from 769 to 289 sec in heavy rain condition. It also reduced the total time for all vehicles from 2490 to 1154 sec.