Yu, Jie (State Key Laboratory of Intelligent Green Vehicle and Mobility, ), Chen, Yihe (Tsinghua University), Zhong, Wenqin (State Key Laboratory of Intelligent Green Vehicle and Mobility, ), Luo, Yugong (Tsinghua University,Beijing)

Hierarchical Decision-Making Method for Vehicle Groups at Edge-Cloud Controlled Intersections under Mixed Traffic Environment

Scheduled for presentation during the Regular Session "Autonomous vehicles - intersection management" (FrAT8), Friday, September 27, 2024, 10:50−11:10, Salon 16

2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC), September 24- 27, 2024, Edmonton, Canada

This information is tentative and subject to change. Compiled on December 26, 2024

Abstract

To improve the safety and traffic efficiency of vehicle groups in mixed traffic systems at intersections, a hierarchical decision-making architecture is proposed. At the upper level, a collaborative decision-making scheme of mixed vehicle formation in road segments based on a prediction of uncertain driving states for Human-Driven Vehicle (HDV) and a parallel optimization strategy for mixed vehicle groups and intelligent traffic signals in intersection area are designed. The control policy permits the mixed vehicle groups meeting certain safety constraints and travel directions constraints to form mixed platoons that pass through the multi-intersection at real-time collaborative driving speeds combined with traffic signals while allowing the traffic signals to adaptively adjust their phase and duration based on real-time decision-making commands. At the lower level, based on the anticipated driving speed obtained from the upper layer, the distributed linear feedback controller is designed for distributed control of Intelligent and Connected Vehicle (ICV) to achieve stable driving. The extensive numerical simulation results show that the proposed strategy can significantly reduce the average travel time based on ensuring multi-vehicle cooperative-driving safety with different volumes and Market Penetration Rate (MPR) of ICVs compared to the benchmark method.