Ant Colony Optimization for Multi-phase Traffic Signal Control

Traffic congestion has significant impact on everyone's daily life. Setting optimal signal sequence and timing at traffic intersections can effectively increase the capacity of existing infrastructures, avoid conflict, and reduce traffic jam. Ant colony optimization (ACO) is a meta-heuristic method based on the behaviors of artificial ants with collaboration and knowledge-sharing mechanism during their food-seeking process. ACO algorithm has been applied to traffic signal optimization in literature; however, current studies often focus on the development of two-phase controllers which is less computationally complex. In this research, we extend the ACO-based approach for eight-phase dual-ring traffic control to reduce vehicle delay and queue length at intersections. Computer simulation results indicate the proposed approach is more efficient than the conventional fully actuated control method for heavy and unbalanced traffic demand.