The Role of Artificial Intelligence and Machine Learning in Federally Supported Surface Transportation 2022 Updates

NHTSA · 2022 · ROSA P / United States. Federal Highway Administration

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This report, published by the Federal Highway Administration (FHWA) in 2022, outlines the role of artificial intelligence (AI) and machine learning (ML) in federally supported surface transportation research. Motivated by the U.S. Department of Transportation’s priority to enhance traffic safety and mobility, the document details how the FHWA’s Exploratory Advanced Research (EAR) Program supports long-term R&D in AI. The program aims to clarify the federal government’s investment strategy in AI, aligning with the National Artificial Intelligence Research and Development Strategic Plan. The report focuses on two primary application areas: leveraging big data to process large-scale transportation datasets and utilizing video analytics to analyze driver behavior and traffic conditions. The EAR Program supports various research initiatives through partnerships with universities, private companies, and public entities. In the domain of big data, current projects include Michigan Technological University’s development of an AI-enhanced framework for autonomous winter road maintenance using deep and reinforcement learning. Carnegie Mellon University is researching predictive real-time traffic management to forecast nonrecurrent incidents, such as accidents or weather events, at least 30 minutes in advance. The University of Cincinnati is exploring cooperative perception and control for freeway operations. Completed projects in this category include the University of California, Davis’s automation of environmental data analysis for wildlife monitoring, PARC’s integration of disparate datasets to enhance traffic safety, and CUBRC’s creation of a dashboard for knowledge discovery in massive transportation datasets. In the field of video analytics, researchers utilize the Strategic Highway Research Program 2 (SHRP 2) Naturalistic Driving Study (NDS) database, which contains 1.2 million hours of video. Current projects include Tufts University and the City College of New York’s development of AI frameworks for highway incident detection, such as wrong-way driving and hazardous objects. A multi-institutional team is building the “Deep InSight” platform to estimate driver states using recurrent neural networks, while the West Big Data Innovation Hub investigates privacy-preserving de-identification methods for video data. The University of Michigan is developing algorithms to classify high-level behaviors, such as cellphone use, and Virginia Tech is creating computer vision methods for automatic annotation of driving situations. Completed projects include Volpe National Transportation Systems Center’s neural networks for detecting roadway features and weather conditions, Carnegie Mellon University’s tools for automated video analysis and driver distraction assessment, and SRI International’s DCode system for comprehensive driver behavior coding. The significance of this research lies in its potential to improve the safety, efficiency, and reliability of the nation’s highway system. By processing vast amounts of data, AI enables better crash detection, predictive traffic management, and infrastructure monitoring. The report emphasizes that future AI research must address challenges related to trustworthiness, ensuring that algorithms are fair, transparent, and free from bias. The FHWA continues to support both dataset analysis and real-time operational tools, aiming to bridge basic academic research with applied industry solutions to benefit the public and support emerging technologies like autonomous vehicles.

Key finding

The FHWA Exploratory Advanced Research Program supports AI and machine learning studies focused on big data analysis and video analytics to enhance highway safety, traffic operations, and driver behavior understanding.

Methodology

review

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• telematics crash prediction

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Methodological Resource: dataset resource, tool software
• Theoretical Contribution: computational model