Fog computing for optimized traffic control strategy

Traffic congestion is a big problem that influences the traffic flow in big cities, so better control of the traffic signals is always searched to solve this type of traffic problems. Fog computing is one of the most efficient paradigms for traffic system control as it enables connecting and analyzing big traffic data to help the control of traffic signals in the appropriate time. There are different optimization methods, which can be used to control traffic signal; one of these is Particle Swarm Optimization (PSO) algorithm, and there is correlation between PSO parameters (particle velocity, position) and traffic parameters (vehicle speed and location). Roundabouts with traffic signals is one of the modern roads infrastructures used to reduce traffic jam. Our objective is to minimize the average delay time in order to decrease the traffic congestion. This paper presents a control strategy called COTSD-PSO for optimizing traffic signaling based on PSO combined with three sub-controllers; this strategy depends on traffic control rules. These sub-controllers are PSO-Jump, PSO-Turn and PSO-Mix depend on two parameters; extension time and urgency degree for the different phases in the traffic cycle. PSO algorithm is applied to optimize the control of the traffic signal network for roundabouts model on fog computing environments using real data from Taif streets in KSA country. The PSO simulation results show that the PSO-Mix has the fastest convergence rate for the optimal solution and the best performance in minimizing the average delay time compared with the other combinations.