Traffic Signal Control Using Machine Learning

Traffic signal has been a long-standing topic in urban traffic control. Ineffective and inflexible traffic control at urban intersections can often lead to an obstruction in traffic flow and will definitely lead to congestion of traffic. How we manage the traffic in a smart manner is a big challenge in urban traffic management. If we introduce new techniques with ways of solving this problem then it will benefit the urban areas. As traffic light systems are everywhere it would be tough to change the system and we also need to cross certain barriers to achieve this task and instead of changing the current system and bringing up new things if we just solve this issue by making some changes in the existing software then it would cut down a lot of other barriers too. All we need to do is just add some algorithms to main software. With recent advances in machine learning, especially reinforcement learning (RL), traffic signal control using advanced machine learning techniques represents a promising solution to tackle this problem. The performance of the proposed method is comprehensively compared with two traditional alternatives for controlling traffic lights. Simulation results indicate that the proposed method significantly reduces the total delay in the network when compared to the alternative methods. Adjacent traffic light intersections will work independently and yet cooperate with each other to a common goal of ensuring the fluency of the traffic flow within traffic network. The experimental results show that the Q-Learning algorithm is able to learn from the dynamic traffic flow and optimized the traffic flow.