The adaptive dynamic programming signal control system for person in a connected vehicle environment

Area of Science:

• Traffic Engineering
• Transportation Systems
• Intelligent Transportation Systems (ITS)

Background:

• Urban traffic congestion and person delay are significant challenges.
• Traditional traffic control methods often focus on vehicles, not individual people.
• Connected Vehicle (CV) data offers new opportunities for traffic management.

Purpose of the Study:

• To propose and evaluate a novel Person-Based Adaptive Control Algorithm (PB-ACA) for urban intersections.
• To minimize average person delay by utilizing CV data and person-level impacts.
• To enhance traffic signal control equity and efficiency in connected environments.

Main Methods:

• Developed a Person-Based Adaptive Control Algorithm (PB-ACA) integrating CV data (occupancy, trajectory, speed).
• Employed a three-layered dynamic programming approach to minimize person delay.
• Integrated a signal phase transition exploration mechanism and generalized vehicle models for accurate platoon prediction.
• Conducted microsimulation experiments using SUMO, comparing PB-ACA against FTCA, ILACA, and VBACVSC.

Main Results:

• PB-ACA reduced average person delay by up to 55% compared to Fixed-Time Control (FTCA).
• PB-ACA achieved 42% reduction compared to Inductive Loop-Actuated Control (ILACA).
• PB-ACA showed an 11% improvement over Vehicle-Based Adaptive CV Signal Control (VBACVSC), particularly for high-occupancy vehicles.

Conclusions:

• PB-ACA effectively minimizes person delay at urban intersections by prioritizing people over vehicles.
• The algorithm demonstrates significant improvements in traffic efficiency and equity.
• PB-ACA offers a promising approach for future intelligent transportation systems leveraging CV data.