Centralized Cooperative Control of Mixed Vehicle Groups Considering Multi-vehicle Influence Under Cloud-Edge-End Collaboration

With the advancement of vehicle-to-everything technology, the scale of mixed vehicle groups composed of connected human-driving vehicles (CHVs) and connected automated vehicles (CAVs) is gradually increasing, which leads to greater difficulty in cooperation and interaction among mixed vehicle groups, thereby affecting the consistency of their driving states. To address this issue, a centralized cooperative control method of mixed vehicle groups considering multi-vehicle influence under cloud-edge-end collaboration is proposed. First, to effectively process and analyze the massive heterogeneous data generated by large-scale mixed vehicle groups, a cloud-edge-end collaborative architecture is designed for mixed traffic scenarios. Then, considering the influence of the states of the two preceding vehicles and the immediate following vehicle on CHVs, as well as the states of all preceding vehicles and the immediate following vehicle on CAVs, the cooperative driving models for CAVs based on molecular dynamics and for CHVs based on fixed weights are established on the cloud control platform, respectively. Furthermore, based on the dynamic influence relationships among mixed vehicle groups, a centralized cooperative control method for mixed vehicle groups is designed based on the cloud-edge-end collaborative architecture, and the consistency conditions for cooperative driving of mixed vehicle groups are obtained using stability and string stability theory. Finally, the comparative simulation experiments are conducted to verify the effectiveness of the proposed method.