Traffic Signal Time Optimization Based on Deep Q-Network

Because cities worldwide have high population concentration, traffic congestion is a key problem that needs to be addressed. As modern technology advances, smart management able collect data from the environment and uses contextual signal assignment determine flow at intersections improve conditions. In this paper, we propose green time allocation system based on deep Q-network (DQN) can maximize capacity assign light according The proposed also aims reduce standard deviation of each lane an intersection by considering waiting time. result, selfish allocations reduced. Thus, achieve better experimental results in dynamic than those phase sequence system.