An Asynchronous Cooperative Signal Control Framework in Urban Road Network

Cooperative signal control in the large-scale urban road network has always been a difficult task due to the non-stationary problem. This study proposes an asynchronous cooperative signal control framework for urban road network to enhance the stability of control in practice. The signal timing plans can be optimized in a parallel fashion in the framework. Signal controllers are able to respond and adapt to the environment change timely through queue prediction based on the asynchronous information from neighbor intersections. Additionally, the rolling-horizon strategy is introduced in the framework to improve the robustness of signal control. Simulation experiments are conducted and the results show the asynchronous framework overperforms the fixed timing plan optimized by Synchro in terms of both stability and motility. The asynchronous cooperative control framework holds huge potential in the real world.