This paper proposes a system that utilizes Vehicle-to-Everything (V2X) technology for autonomous vehicles to exchange data and make real-time decisions. The system also operates a federated learning scheme to create a predictive model of the transportation network, improving learning performance and ensuring privacy. The system is evaluated using a non-IID (independent and identically distributed) dataset, as real-world autonomous vehicle data is not yet available.
Publication date: 26 Jan 2024
Project Page: Not provided
Paper: https://arxiv.org/pdf/2401.13848
