Work

Deep Multi-Agent Reinforcement Learning for Real-World Signalized Traffic Corridor Control

Year: 2024

Type: conference-paper

Source: International Conference on Machine Learning and Applications

DOI: 10.1109/icmla61862.2024.00093

Authors: Salman Shuvo, Sayak Mukherjee, Samrat Chatterjee, Sonja Glavaski, Draguna Vrabie, Geline Canayon, Matthew Juckes, Raimundo Rodulfo

Institution: Pacific Northwest National Laboratory

Language: English

Topic: Traffic control and management

Subfield: Control and Systems Engineering

Field: Engineering

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: closed

Funder: U.S. Department of Energy

Award: TRAFFIC TECHNOLOGY SERVICES, INC. NEW OPEN 2021 AWARD CONTROL NUMBER: 2459-2552 TITLE: AUTONOMOUS INTELLIGENT ASSISTANT (AUTONOMIA): RESILIENT AND ENERGY-EFFICIENT CITY-WIDE TRANSPORTATION OPERATIONS PACIFIC NORTHWEST NATIONAL LABORATORY, THE UNIVERSITY OF CALIFORNIA AT BERKELEY, AIMSUN, SIEMENS TECHNOLOGY, TRAFFIC TECHNOLOGY SERVICES, AND THE CITY OF CORAL GABLES, FL, WILL COMBINE ARTIFICIAL INTELLIGENCE WITH MULTISCALE SIMULATION AND REAL-TIME CONTROL TO REDUCE CONGESTION, IMPROVE ENERGY EFFICIENCY, AND REDUCE CO2 EMISSIONS ACROSS REGIONAL TRANSPORTATION SYSTEMS. AUTONOMIA’S TRAFFIC FORECASTING AND REAL-TIME MANAGEMENT APPROACH WILL LEVERAGE EMERGENT CONNECTED AUTOMATED VEHICLES WITH VARIED LEVELS OF AUTONOMY (CAVS), SENSING, AND SIGNALING TECHNOLOGIES WITHIN A UNIFIED LEARNING, SIMULATION, AND CONTROL PARADIGM TO IMPROVE ENERGY EFFICIENCY BY MORE THAN 20% AND TO REDUCE EMISSION BY AT LEAST 20%.

OpenAlex ID: W4408146311

FWCI: 1.373

Cites: 19

Cited by: 2

Related to: 10