Research by the POLIMI IoTLab, developed in collaboration with the National Research Council of Italy (CNR), has been recognised for its contribution to the future of connected and automated mobility.

The POLIMI School of Management at Politecnico di Milano has received the 2024 Best Paper Award from the Elsevier journal Vehicular Communications for the paper “Decentralized Federated Learning for Extended Sensing in 6G Connected Vehicles”, developed by the POLIMI IoTLab in collaboration with the National Research Council of Italy (CNR).
Vehicular Communications is a leading international peer-reviewed journal focusing on all aspects of vehicle communications, including vehicle-to-vehicle and vehicle-to-infrastructure interactions, and represents a key reference point for research on intelligent and connected mobility.

The paper was authored by Luca Barbieri, former POLIMI IoTLab postdoctoral researcher and currently Research Scientist at Nokia Bell Labs, Stuttgart, Germany, Stefano Savazzi, Senior Researcher at the Institute of Electronics, Computer and Telecommunication Engineering (IEIIT) of CNR, Mattia Brambilla, Assistant Professor of the IoTLab, with the  Department of Electronics, Information and Bioengineering (DEIB), Politecnico di Milano, and Monica Nicoli, Associate Professor at the POLIMI School of Management and coordinator of the research at the IoTLab.

Published in January 2022 in the special issue “Revolutionary Paradigms for Smart Connected Vehicles in the 6G Era”, the paper addresses the growing role of connected vehicles, vehicles equipped with sensing and communication technologies that enable them to exchange information in real time with other nearby vehicles and with the road infrastructure.

The study investigates how Federated Learning, a collaborative approach to machine learning, can enhance the perception capabilities of connected and automated vehicles. Instead of sharing raw sensor data, each vehicle trains a local model and exchanges only model parameters through vehicle-to-everything (V2X) communications, improving data privacy while reducing communication loads.

This approach allows vehicles to collectively improve the interpretation of information gathered by on-board sensors, such as lidar (Light Detection and Ranging – a remote sensing technology that uses pulsed laser light to measure distances and create a precise 3D model of the sensed area), enabling more accurate detection and classification of road users and objects, even beyond the vehicle’s direct field of view. By combining Federated Learning with next-generation 6G communication technologies, the research highlights the potential for ultra-low latency, highly reliable connectivity, supporting cooperative driving services and contributing to safer and more sustainable road mobility.

The achievement underscores the School’s commitment to fostering interdisciplinary approaches that support innovation and the evolution of connected and automated mobility.

Click here to view the certificate for the award: VEHCOMBestPaperAward2025