Density-Based Optimization Approach for Coordinated Traffic Management on a Simplified Traffic Model using Genetic Algorithms

The optimization of intelligent traffic management systems is a computationally complex task and therefore requires algorithms that can efficiently find the optimal solutions in a vast solution space in a short amount of time. One such algorithm is the genetic algorithm which uses biological evolution inspired operations to find the global minima of a fitness function. This paper proposes a genetic algorithm-based control scheme for an ITMS composed of multiple smart traffic light modules implemented thru MATLAB. It was found that the density-based genetic algorithm system improved the time to route 1000 vehicles across the road network by 22.8% in comparison with a passive system. The genetic algorithm controller was able to dynamically change the next state of each intersection to optimize traffic flow throughout 4 intersections. Overall, the system is a viable control method alternative for traffic light systems and can effectively reduce traffic congestion and increase vehicle throughput.