Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Traffic Signal Optimization by Integrating Reinforcement Learning and Digital Twins

Conference · · 2023 IEEE Smart World Congress (SWC)
 [1];  [2];  [3];  [3];  [4];  [4];  [3]
  1. University of Tennessee at Chattanooga,Department of Computer Science and Engineering,USA; University of Tennessee at Chattanooga
  2. Oak Ridge National Laboratory,Applied Research for Mobility Systems,USA
  3. University of Tennessee at Chattanooga,Department of Computer Science and Engineering,USA
  4. Georgia Institute of Technology,School of Civil and Environmental Engineering,USA

Machine learning (ML) methods, especially reinforcement learning (RL), have been widely considered for traffic signal optimization in intelligent transportation systems. Most of these ML methods are centralized, lacking in scalability and adaptability in large traffic networks. Further, it is challenging to train such ML models due to the lack of training platforms and/or the cost of deploying and training in a real traffic networks. This paper presents an approach for the integration of decentralized graph-based multi-agent reinforcement learning (DGMARL) with a Digital Twin (DT) to optimize traffic signals for the reduction of traffic congestion and network-wide fuel consumption related to stopping. Specifically, the DGMARL agents learn traffic state patterns and make decisions regarding traffic signal control with assistance from a Digital Twin module, which simulates and replicates the traffic behaviors of a real traffic network. The proposed approach was evaluated using PTV-Vissim [1], a microscopic traffic simulation platform. PTV-Vissim is also the simulation engine of the DT, enabling emulation and optimization of the traffic signals on the MLK Smart Corridor in Chattanooga, Tennessee. Compared to an actuated signal control baseline approach, experiment results show that Eco_PI, a developed performance measure capturing the impact of stops on fuel consumption, was reduced by 44.27% in a 24-hour and an average of 29.88% in a PM-peak-hour scenario.

Research Organization:
University of Tennessee at Chattanooga
Sponsoring Organization:
USDOE Office of Energy Efficiency and Renewable Energy (EERE)
DOE Contract Number:
EE0009208
OSTI ID:
2439938
Journal Information:
2023 IEEE Smart World Congress (SWC), Journal Name: 2023 IEEE Smart World Congress (SWC)
Country of Publication:
United States
Language:
English

References (16)

Digital Twin Technologies and Smart Cities book January 2020
STMARL: A Spatio-Temporal Multi-Agent Reinforcement Learning Approach for Cooperative Traffic Light Control journal June 2022
MLK Smart Corridor: An Urban Testbed for Smart City Applications conference December 2019
Deep LSTM Recurrent Neural Networks for Arterial Traffic Volume Data Imputation journal April 2021
Adaptive Digital Twin and Multiagent Deep Reinforcement Learning for Vehicular Edge Computing and Networks journal February 2022
Digital Twins in the Intelligent Transport Systems journal January 2021
Real-Time Traffic Simulation With a Microscopic Model journal December 2006
Fuel Consumption Intersection Control Performance Index journal April 2021
Spatio-Temporal Graph Dual-Attention Network for Multi-Agent Prediction and Tracking journal August 2022
Digital Twin in Industry: State-of-the-Art journal April 2019
Digital twins and artificial intelligence in transportation infrastructure: Classification, application, and future research directions journal July 2022
Investigating impacts of various operational conditions on fuel consumption and stop penalty at signalized intersections journal December 2022
Digital Twin: Values, Challenges and Enablers From a Modeling Perspective journal January 2020
Evolutionary reinforcement learning multi-agents system for intelligent traffic light control: new approach and case of study journal October 2022
Development of a Connected Corridor Real-Time Data-Driven Traffic Digital Twin Simulation Model journal December 2021
Field-Based Prediction Models for Stop Penalty in Traffic Signal Timing Optimization journal November 2021