Agent-Based Simulation of Ecospeed-Controlled Vehicles at Signalized Intersections

Abstract

Ecospeed control systems attempt to reduce vehicle fuel consumption levels by optimization of vehicle trajectories in the vicinity of signalized intersections while these systems account for traffic signal timing constraints. The proposed algorithm uses dynamic programming to compute the vehicle trajectory needed to navigate through the intersection with the amount of minimum fuel subject to several constraints, including vehicular interactions, traffic signal timing changes, and vehicle and roadway constraints. The proposed application uses infrastructure-to-vehicle and vehicle-to-vehicle communication to receive traffic signal and vehicle data. The research presented in the paper developed an agent-based modeling tool to simulate and test the system under various traffic volume and market penetration levels. The simulation model used a variety of microscopic inputs, such as the roadway vertical profile, roadway surface condition, traffic volumes, and traffic signal timing information. The system was tested on a sample signalized intersection and produced fuel consumption reductions of 30% and travel speed increases of 200%, on average, within the vicinity of the intersection. Actual savings in total trip time, average speed, and the amount of fuel consumed depend on the trip profile, including the number of intersections and total trip length.