Resource-Aware Cooperative Traffic Control in IoV-Enhanced Transportation Systems

With the development of the Internet of Vehicles (IoV), Intelligent Connected Vehicles (ICVs) are set to replace human-driven vehicles as the backbone of future transportation systems. The implementation of IoV-enhanced transportation systems relies heavily on communication and computing resources. To address the gap in previous studies on the relationship between resource networks and transportation systems, we propose a bidirectional interaction model with safety distance as an intermediary and introduce a joint optimization problem for travel efficiency and Quality of Service. The analysis of this model reveals that the driving speed of ICVs plays a crucial role in optimizing traffic efficiency. We propose the integration of variable speed limits with intelligent traffic signal control to overcome the limitations of traditional traffic signal management strategies. Furthermore, we divide the road network into basic units consisting of target and adjacent intersections, and introduce a neighborhood cooperation multi-agent traffic control algorithm that leverages an attention mechanism to capture and learn relationships, thereby enhancing agent collaboration. The simulation results demonstrate that our algorithm outperforms other methods in terms of road throughput, the number of waiting vehicles, and intersection pressure, while overcoming the dimensional explosion of centralized algorithms and the local optimization issues in distributed algorithms.