Neural Infrastructure a Federated AI Framework for Predictive Resilience in U.S. Transportation Systems
Year: 2025
Type: article
Abstract: Because autonomous vehicles and intelligent traffic systems play a bigger role, robust, scalable and secure traffic prediction models are more necessary than ever. The integration of federated AI is studied in this article as a leading way to improve vehicle upkeep and guide traffic in autonomous transportation. Using just the most current research in deep learning for traffic flow modeling, forecasting vehicle routes and decision-making algorithms, the authors highlight that using federated AI can enhance safety, cut down on costs for repairs and increase up-to-date knowledge of how things are running, while preserving privacy. When learning is spread among connected vehicles and support nodes, the suggested method can support both geographical adaptability and nationwide scalability. As a result of this shift, everything from the design to the development of AI-driven mobility becomes smoother and people gain more trust. The analysis of present predictive methods and the issues of using them shows that federated AI holds great promise for intelligent transportation systems.
Authors: Muhammad Faheem, Muhammad Awais, Aqib Iqbal, Hasnain Zia
Institutions: Cumberland University, COMSATS University Islamabad
Language: English
Subfield: Building and Construction
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4411285934