AI-Driven Speed Optimization for Mixed Traffic Networks Using Adaptive Learning
TL;DR
This study introduces an AI-based speed control mechanism leveraging deep reinforcement learning to enhance freeway traffic conditions under varying levels of autonomous vehicle integration to provide a foundation for future adaptive traffic management strategies in evolving transportation ecosystems.
Abstract
Managing traffic flow efficiently in mixed vehicle environments remains a critical challenge in modern transportation systems. This study introduces an AI-based speed control mechanism leveraging deep reinforcement learning to enhance freeway traffic conditions under varying levels of autonomous vehicle integration. Using a data-driven simulation approach, the model dynamically adjusts speed limits based on real-time traffic observations, optimizing throughput and minimizing congestion. The proposed system is tested in a high-fidelity simulated environment, demonstrating improved efficiency at moderate levels of autonomous vehicle deployment. However, results indicate that excessive reliance on automated control can reduce the effectiveness of dynamic speed interventions. These insights provide a foundation for future adaptive traffic management strategies in evolving transportation ecosystems.