Actuated traffic signal control has been widely used to accommodate varying arrivals at signalized intersections. In practice, however, primary control parameters (e.g., minimum and maximum green times) in actuated controllers are roughly predefined on the basis of the limited sampling traffic data, which actually cannot well respond to the real-time traffic demands in operation. To address this issue, this paper presents an innovative algorithm to dynamically determine the control parameters against varying traffic conditions measured by the License Plate Recognition (LPR) data which is detected by Electronic Police devices. With this algorithm, the vehicle rates detected from the LPR datasets at the upstream intersection are firstly used to predict the arrivals for the connected approach of the downstream intersection. Then, the signal timing parameters of the downstream intersection are determined based on this prediction. The control parameters are updated for the next cycle based on the new arrivals and this process will be repeated in iterative calculations with an attempt to match the real-time traffic demands. The optimization control objective is two-fold: to maximize the total intersection traffic throughputs while minimizing the overall delay. The performance of the proposed control algorithm has been evaluated using the microscopic simulation method. The performance comparison of the improved actuated signal control scheme with the existing pre-timed control scheme indicates an obvious improvement by the developed algorithm based on the LPR data.

CICTP