Adaptive Traffic Control Algorithm Based on Back-Pressure and Q-Learning

Nowadays traffic congestion has increasingly been a significant problem, which results in longer travel time and aggravates air pollution. Available work showed that backpressure based traffic control algorithms can effectively reduce traffic congestion. However, those work control traffic based on either inaccurate traffic information or local traffic information, which causes inefficient traffic scheduling. In this paper, we propose an adaptive traffic control algorithm based on backpressure and Q-learning, which can efficiently reduce congestion. Our algorithm controls traffic based on accurate real-time traffic information and global traffic information learned by Q-learning. As verified by simulation, our algorithm significantly decreases average vehicle traveling time from 16% to 36% when compared with state-of-the-art algorithm under tested scenarios.