A method for controlling a plurality of traffic intersections

Abstract

The present invention relates to a method (100) for controlling a plurality of traffic intersections, characterised by the steps of obtaining general traffic data of each intersection by each camera coupled with a deep learning model (101), wherein the general traffic data comprises queue, residual, and occupancy of each transaction, detecting and classifying each vehicle at a first intersection by a first camera (102), assigning a unique identifier to the detected vehicle for tracking purposes (103), identifying the detected vehicle at a subsequent intersection based on the unique identifier by a second camera (104), calculating individual data including journey time, waiting time, and speed of the detected vehicle (105), delivering the general traffic data of each intersection and the individual data of each vehicle to a remote processing unit over a network (106), generating a traffic signal timing plan for each intersection based on the delivered data using an adaptive algorithm and historical intersection data by the remote processing unit (107), wherein the traffic signal timing plan comprises offset, split and cycle length of the traffic signal, and delivering the traffic signal timing plan to a traffic light controller over the network (108). Drawing accompanying abstract: Figure 1