Coordinated Traffic Flow Control in a Connected Environment

The current traffic system operates most of the time as an open loop dynamical system with control actions limited to traffic light control at intersections of arterial streets and ramps to highways. These control actions rely on limited sensor data to make decisions which are often far from the optimum. In the case of ramps, the control metering switches off when the ramp queue exceeds certain capacity which takes place usually during peak times when control is needed the most. As the traffic demand increases the current situation is not sustainable as congestion will get worse and inefficiencies increase unless new sensor technologies and traffic flow control techniques are developed.

The deployment of vehicle to infrastructure (V2I) and vehicle to vehicle (V2V) technologies is the inevitable choice that will open the way to a revolution in traffic flow control and optimization as connectivity will provide real time traffic information that is necessary for the design of accurate traffic control techniques to better control traffic and prevent the onset of congestion by balancing traffic demand across the network.

This project will investigate how connectivity provided by such communication technologies can be used to develop traffic flow control systems that will enhance mobility and safety, reduce queues at ramps with positive benefits to transportation efficiency and environment. The focus of the project is to develop a COoRdinated traffic FLOw control system (CORFLO) that will involve variable speed limit, lane change, ramp metering control that also take into account traffic light control in arterial intersections in order to improve mobility, reduce queues at ramps in a way that the ramp metering control is not switched off as it is currently done. Despite recent research efforts in these different control strategies there is not much research in coordinating all relevant control strategies to meet an overall objective for a transportation network whose different parts are currently controlled independently and in an inefficient way. Lane change and variable speed limit control for example could easily help the ramp metering controller maintain a flow without exceeding the bounds of ramp queues whereas traffic light control of intersections feeding the ramp can also contribute by controlling the inflow to the ramp in an intelligent way. The objective of the CORFLO control scheme is to balance the load of the traffic demand across the network so that situations where some links or ramps are congested whereas adjacent ones are operating below capacity are avoided. This can be achieved by exercising the following control actions: Variable speed limit(VSL), lane change(LC), ramp metering (RM) and traffic light control in a coordinated manner in order to deal with normal traffic conditions, accidents, bottlenecks and high demand situations. Due to time and budget constraints the first version of CORFLO will focus on a coordinated design of VSL, LC and RM that takes into account the timing of traffic lights that affect the highway ramps as well as the ramp queue constraints. It will incorporate a module that allows the incorporation of different traffic light algorithms which will coordinate with the highway traffic and ramp metering controllers.

The microscopic simulator developed under previous projects to simulate the traffic of large road network in the Los Angeles Metropolitan Area will be used to evaluate CENCOS whereas the EPA model MOVES will be used to analyze emissions and impact on the environment. Part of the network has a high volume of trucks leading to nonhomogeneous vehicular traffic. The impact of trucks and larger size vehicles in general on traffic flow and how the CORFLO control strategies will take into account different classes of vehicles will be also investigated.

Research Area

Tags