Asyn-Light: Asynchronous Traffic Signal Cooperative Control Through Spatial-Temporal Transformer

Recent multi-agent deep reinforcement learning (MADRL) approaches have shown notable benefits in multi-intersection traffic signal control (M-TSC). However, spatial-temporal coupling and action heterogeneity are often overlooked. Interactions among traffic states and signal actions introduce complex coupling and hysteresis across space and time, while local actions vary due to fluctuating flows. We propose Asyn-Light, a MADRL-based M-TSC model that captures spatial-temporal traffic features and leverages action repetition for asynchronous cooperative control. To address spatial-temporal coupling, Asyn-Light employs a feature extraction framework based on a spatial-temporal Transformer with a stacked spatial graph module. To handle action heterogeneity, it uses a repetition-enabled MADRL to generate asynchronous multi-step policies for each intersection. We evaluate Asyn-Light on both synthetic and real-world datasets, analyzing each component’s contribution. Results show that Asyn-Light consistently outperforms baselines in diverse settings, effectively balancing action smoothness with policy adaptability.