Due to the rising of traffic volumes, city expressways are experiencing significant congestion, leading cities to introduce parallel surface roads as alternative routes to help drivers bypass traffic bottlenecks. In such scenarios, road pricing on expressways and traffic signal controls on surface roads have shown effectiveness in alleviating citywide congestion. However, existing research on these strategies often neglects the optimization of the entire parallel road network, failing to simultaneously address congestion on both expressways and surface roads. In this paper, we propose a Collaborative Optimization mechanism of Price and Traffic Signal Control (CO-PTSC) based on deep reinforcement learning, a novel approach that integrates road pricing and traffic signal control, using deep reinforcement learning to optimize traffic flow and minimize travel times across the parallel road network. Our experimental results demonstrate significant improvements in network efficiency and reduced travel times.

Road pricing is an urban traffic management mechanism to reduce traffic congestion. Currently, most of the road pricing systems based on predefined charging tolls fail to consider the dynamics of urban traffic flows and travelers’ demands on the arrival time. In this paper, we propose a method to dynamically adjust online road toll based on traffic conditions and travelers’ demands to resolve the above-mentioned problems. The method, based on deep reinforcement learning, automatically allocates the optimal toll for each road during peak hours and guides vehicles to roads with lower toll charges. Moreover, it further considers travelers’ demands to ensure that more vehicles arrive at their destinations before their estimated arrival time. Our method can increase the traffic volume effectively, as compared to the existing static mechanisms.