Traffic light control with reinforcement learning

Abstract

Urban traffic signal optimization is important for alleviating congestion in urban transportation systems. This study proposes a real-time traffic light control algorithm based on deep Q learning with a reward function that accounts for queue lengths, delays, travel times, and throughput. The model dynamically decides phase changes based on current traffic conditions. The training of the deep Q network involves an offline stage from pre-generated data with fixed signal timing and an online stage using real-time traffic data. A deep Q network structure with aphase gate component is used to simplify the model's learning task under different phases. Amemory palace" mechanism is used to address sample imbalance during the training process. Both synthetic and real-world traffic flow data are used to validate our approach under an urban road intersection scenario in Hangzhou, China. Results demonstrate significant performance improvements of the proposed method in reducing vehicle waiting time (57.1% to 100%), queue lengths (40.9% to 100%), and total travel time (16.8% to 68.0%) compared to traditional fixed signal timing plans.