Digital Twin and Agentic AI Framework for Real-Time Traffic Signal Control

A novel approach to optimizing traffic signals employs a digital twin of transportation infrastructure overseen by autonomous AI for immediate decision-making. This system utilizes physical sensors and edge computing to gather live traffic data and continuously update the digital twin for flow simulation. Traffic signals are adjusted automatically in response to congestion, delays, and traffic patterns. The architecture consists of three layers: perception (data acquisition from physical systems), conceptualization (processing through LangChain), and action (connecting to Model Context Protocol and traffic management APIs). Findings indicate a reduction in waiting times at traffic signals.