Urban intelligent traffic signal coordination control system based on machine learning

In order to improve the traffic capacity of the intersection and shorten the traffic delay, a new urban intelligent traffic signal coordination control system based on machine learning is proposed in this paper. The hardware of the system is designed, including annunciator, vehicle detection module, traffic flow statistics module and signal control module. The reinforcement learning algorithm in machine learning is used to train and learn the traffic vehicle data, so as to extract the spatial characteristics of the traffic environment state. Under the constraint of the minimum loss function, the training objective is solved and the output result of the optimal coordinated control of traffic signals is output. The experimental results show that compared with the traditional control system, the maximum traffic capacity of the intersection under the control of this system is 2400 vehicles/hour, and the maximum traffic delay is no more than 30s.