Adaptive Traffic Signal Timing Optimization Using Deep Reinforcement Learning in Urban Networks

Zhonghao Wu; Shaobo Wang; Chunhe Ni; Jiang Wu

doi:10.69987/AIMLR.2024.50405

Authors

Zhonghao Wu Computer Engineering, New York University, NY, USA Author
Shaobo Wang Computer Science and Engineering, Santa Clara University, CA, USA Author
Chunhe Ni Computer Science, University of Texas at Dallas, Richardson, TX, USA Author
Jiang Wu Computer Science, University of Southern California, Los Angeles, CA, USA Author

DOI:

https://doi.org/10.69987/AIMLR.2024.50405

Keywords:

Traffic Signal Control, Deep Reinforcement Learning, Deep Deterministic Policy Gradient, Adaptive Control Systems

Abstract

This document please leave the flashlight that uses the extra work (DRL) for local communication. LEFTMENT Activity GRARIAR (DDPG) MATAL-Multi-multi-multi-multisgre traffic IET, QUEUE LONGS, and Distance Models. Problems the management of the controller is designed to continue with the operating sites including reductions, by maintenance operations. The Network Network architecture is designed, featuring special components and mental health regulations from cars. Procedures are used and measured two test platform and real car information from the main city of metrozopolitan from 12 intersections. Experiment that the recommended suggestions have enhanced achievement processes existing, including the suspension of 23.5% in the average attack. The latest-world of validity is completed in a 6-month-monthly basis. Using the use of electric wiring for local work and clouds managed cooperation, improve the intelligence application.