About Traffic AI Systems

Traffic AI Systems is an open-access platform providing AI-powered analytical tools for transportation engineers, urban planners, and public agencies across the United States.

The platform bridges a recognized gap between federal transportation research priorities and the tools available to practicing professionals. Its tools address core analytical challenges in infrastructure safety assessment, travel demand analysis, pedestrian circulation modeling, traffic flow visualization, and AI-optimized signal control — using nationally standardized public data sources including the U.S. Census Bureau, OpenStreetMap, and federal crash records.

Available Tools:

AI Infrastructure Readiness Advisor. Evaluates intersection-level autonomous vehicle readiness across eight infrastructure dimensions, with scoring calibrated to SAE automation levels (L2–L5) and AI-generated engineering recommendations. Addresses the FHWA-identified need for systematic AV infrastructure assessment.
AI Traffic Signal Optimizer. Digital twin microscopic simulation of real-world traffic networks with AI-controlled dynamic signal timing optimization to reduce congestion in real time.
Commute OD & Modal Split Generator. Automated retrieval and normalization of CTPP origin-destination flows, ACS modal splits, and auto ownership data from the U.S. Census Bureau API.
Pedestrian Assignment Simulator. Graph-based sidewalk assignment using A* pathfinding with direction-invariant volume aggregation on OSM walk networks.
Traffic Portal Generator. Automated JTW/RJTW traffic flow analysis with real-world routing via Google Directions API.

All tools are publicly accessible at no cost and designed to work in any U.S. jurisdiction.

Developed independently by Yihang Zhu.