Traffic signal control for connected and non-connected vehicles

Transportation is one of the city's processes to achieve an efficient, livable and sustainable community aiming to improve the citizens' quality of life and the perception of smartness of a city. Each of us experiences unnecessary delays and expects improvement in mobility with the introduction of new technologies, specially in traffic signal control due to its big influence in urban traffic networks. One trend in modern vehicle technologies is that Connected Vehicles (CV) will communicate to the traffic infrastructure, so called V2I (Vehicle-to-Infrastructure), which may enable cities to provide better services through cooperation under limited infrastructure. Therefore, this paper proposes and evaluates three algorithms for traffic control using connected vehicles instead of stationary detectors: (i) a Dynamic Maximum Gap (DMG) between arrivals at stop line (specific for each vehicle), and (ii) the Throughput Adjusted Delay (TAD) accounting the relation between intersection throughput and delay, while (iii) the Throughput Adjusted Stopped Time (TAST) uses stopped time (waiting time) for such relation instead of delay. We demonstrate that our DMG algorithm at non-peak flows, compared to traditional actuated control, reduces the travel time up to 15%, waiting time (time spent with speed lower than 0.1 m/s) by almost 80%, and delay 50%. The TAD and TAST strategies maintain good performance even at 10% penetration rate of CV. Future research will contribute to the assessment of the environmental and economical benefits of traffic signal control using CVs, implementation on more realistic scenarios, as well as exploration of other information from CVs and application of advices sent from the infrastructure to vehicles.