Publisher

University of Tennessee at Chattanooga

Place of Publication

Chattanooga (Tenn.)

Abstract

In urban transportation systems, integrating pedestrian movements into traffic signal timing optimization is crucial for smooth and safe traffic flow. This study investigates the Decentralized Graph-based Multi-Agent Reinforcement Learning (DGMARL) method for signal timing optimization, considering both pedestrian and vehicle traffic states. Evaluating fixed and adaptive pedestrian accommodation strategies, the study assesses their impact on traffic flow and signal timing. Using a Digital Twin microscopic traffic simulation model of the MLK Smart Corridor in Chattanooga, Tennessee, the approach's effectiveness is analyzed. Results show that the strategy of signal timing optimization with adaptive pedestrian request improves Eco_PI by significantly compared to fixed pedestrian recall methods.

Document Type

presentations

Language

English

Rights

http://rightsstatements.org/vocab/InC/1.0/

License

http://creativecommons.org/licenses/by/4.0/

Share

COinS
 

Pedestrian-Involved Traffic Signal Optimization Using Decentralized Graph-based Multi-Agent Reinforcement Learning

In urban transportation systems, integrating pedestrian movements into traffic signal timing optimization is crucial for smooth and safe traffic flow. This study investigates the Decentralized Graph-based Multi-Agent Reinforcement Learning (DGMARL) method for signal timing optimization, considering both pedestrian and vehicle traffic states. Evaluating fixed and adaptive pedestrian accommodation strategies, the study assesses their impact on traffic flow and signal timing. Using a Digital Twin microscopic traffic simulation model of the MLK Smart Corridor in Chattanooga, Tennessee, the approach's effectiveness is analyzed. Results show that the strategy of signal timing optimization with adaptive pedestrian request improves Eco_PI by significantly compared to fixed pedestrian recall methods.