Stabilizing Traffic via Autonomous Vehicles: A Continuum Mean Field Game Approach

Abstract

This paper presents scalable traffic stability analysis for both pure autonomous vehicle (AV) traffic and mixed traffic based on continuum traffic flow models. Human vehicles are modeled by a non-equilibrium traffic flow model, i.e., Aw-Rascle-Zhang (ARZ), which is unstable. AVs are modeled by the mean field game which assumes AVs are rational agents with anticipation capacities. It is shown from linear stability analysis and numerical experiments that AVs help stabilize the traffic. Further, we quantify the impact of AV's penetration rate and controller design on the traffic stability. The results may provide insights for AV manufacturers and city planners.

arXiv comment6 pages

References in corpus (5)

• Dissipation of stop-and-go waves via control of autonomous vehicles: Field experiments
• Mean Field Multi-Agent Reinforcement Learning
• Mistuning-based Control Design to Improve Closed-Loop Stability of Vehicular Platoons
• Constructing set-valued fundamental diagrams from jamiton solutions in second order traffic models
• A Game-Theoretic Framework for Autonomous Vehicles Velocity Control: Bridging Microscopic Differential Games and Macroscopic Mean Field Games

Cited by in corpus (7)

• A Physics-Informed Deep Learning Paradigm for Car-Following Models
• Physics-Informed Deep Learning For Traffic State Estimation: A Survey and the Outlook
• Dynamic driving and routing games for autonomous vehicles on networks: A mean field game approach
• A Game-Theoretic Framework for Autonomous Vehicles Velocity Control: Bridging Microscopic Differential Games and Macroscopic Mean Field Games
• Liability Design for Autonomous Vehicles and Human-Driven Vehicles: A Hierarchical Game-Theoretic Approach
• A Survey on Autonomous Vehicle Control in the Era of Mixed-Autonomy: From Physics-Based to AI-Guided Driving Policy Learning
• Determining state space anomalies in mean field games