Simulation-Based Bidirectional Control between AVs and Adaptive Traffic Signal Control Using Real-World V2X-Infrastructure

Abstract

This research addresses the cooperative optimization challenge in mixed-traffic urban intersections by designing a bidirectional coordination framework between connected autonomous vehicles (AVs) and adaptive traffic signal controllers. Leveraging real-world V2X infrastructure at the Mobility Innovation Campus (MIC) in Munich, we constructed a high-fidelity microscopic simulation environment in SUMO/TraCI. We proposed a Deep Q-Network (DQN) reinforcement learning algorithm for dynamic phase optimization and evaluated intersection delay, queue length, and throughput under varying demand and V2X penetration rates.

Keywords: Autonomous Vehicles, V2X, SUMO/TraCI, Reinforcement Learning, DQN, Adaptive Signal Control.

Recommended citation: Xu, An. (2026). "Simulation-Based Bidirectional Control between AVs and Adaptive Traffic Signal Control Using Real-World V2X-Infrastructure." Master's Thesis, Technical University of Munich Asia.