A macroscopic Game-Theoretic Model Predictive Control framework for emergency vehicle prioritization in urban traffic...

Area of Science:

• Intelligent Transportation Systems
• Traffic Engineering
• Control Theory

Background:

• Urban traffic congestion poses challenges for emergency vehicle (EV) response.
• Traditional traffic signal preemption can cause secondary congestion and network instability.

Purpose of the Study:

• To propose a decentralized Game-Theoretic Model Predictive Control (GTMPC) framework for efficient EV prioritization.
• To maintain traffic equilibrium while facilitating rapid EV movement in urban networks.

Main Methods:

• Developed a decentralized, macroscopic GTMPC framework integrating a weighted non-cooperative game with Model Predictive Control (MPC).
• Each intersection acts as an autonomous agent strategically allocating green-time using dynamic priority weights.
• Validated the framework using SUMO simulations with real-world traffic data from Dublin, Ireland.

Main Results:

• GTMPC significantly reduced Average Suffering Red-lights, Average Time Loss, and Average Stop Count compared to Fixed-Time and Classical MPC.
• Achieved substantial improvements in Average Speed for emergency vehicles.
• Under heavy traffic, GTMPC reduced Average Waiting Time by over 20% and maintained robust performance.

Conclusions:

• The proposed GTMPC framework offers an efficient and adaptive solution for emergency vehicle prioritization in urban traffic.
• Its decentralized structure demonstrates scalability and robustness for real-time traffic control applications.
• GTMPC effectively manages concurrent EV arrivals through strategic multi-player negotiations.