PRINCE: A Proactive Client Selection in Federated Learning for Connected and Autonomous Vehicles

• Amanda Lopes UFPA
• John Sousa UFPA
• Lucas Bastos UNIFESPA
• Lucas Pacheco UFPA
• Iago Medeiros UFPA
• Denis Rósario UFPA
• Eduardo Cerqueira UFPA

DOI: https://doi.org/10.5753/sbrc.2026.19940

Resumo

Federated Learning (FL) enables cooperative training among Connected and Autonomous Vehicles (CAVs) while preserving data privacy. However, the volatility of vehicular environments, characterized by frequent link interruptions and high mobility, poses a significant obstacle to system robustness, often leading to client failures (e.g., connection, resource, aborts) that degrade global model performance. In this paper, we introduce PRINCE (Proactive Reliability-driven INtelligent Client sElection), a framework that integrates stochastic mobility modeling directly into the FL decision-making loop. In its operation, PRINCE synergizes Shannon Entropy to quantify the informational value of local data with a probabilistic mobility model to proactively filter unstable nodes before selection. Evaluation results demonstrate that PRINCE achieves a final accuracy of 83.90% and a training success rate of 61.32%. Crucially, our approach outperforms state-of-the-art reactive baselines, delivering gains of up to 9.22% in accuracy and a 3.5× improvement in resource efficiency.

Referências

Chellapandi, V. P., Yuan, L., Brinton, C. G., Żak, S. H., and Wang, Z. (2023). Federated learning for connected and automated vehicles: A survey of existing approaches and challenges. IEEE Transactions on Intelligent Vehicles, 9(1):119–137.

Chen, D., Deng, T., Jia, J., Feng, S., and Yuan, D. (2025). Mobility-aware decentralized federated learning with joint optimization of local iteration and leader selection for vehicular networks. Computer Networks, 263:111232.

Chen, H. and Vikalo, H. (2024). Heterogeneity-guided client sampling: Towards fast and efficient non-iid federated learning. Advances in Neural Information Processing Systems, 37:65525–65561.

de Souza, A. M., Maciel, F., da Costa, J. B., Bittencourt, L. F., Cerqueira, E., Loureiro, A. A., and Villas, L. A. (2024). Adaptive client selection with personalization for communication efficient Federated Learning. Ad Hoc Networks, 157:103462.

Deva Hema, D. and Rajeeth Jaison, T. (2024). Efficient collision risk prediction model for autonomous vehicle using novel optimized lstm based deep learning framework. International Journal of Intelligent Transportation Systems Research, 22(2):352–362.

Elbir, A. M., Soner, B., Çöleri, S., Gündüz, D., and Bennis, M. (2022). Federated learning in vehicular networks. In IEEE International Mediterranean Conference on Communications and Networking (MeditCom), pages 72–77.

Guo, X., Zhao, C., and Wang, Y. (2019). Traffic sign recognition based on joint convolutional neural network model. the 2nd International Conference on Big Data Technologies (ICBDT ’19), page 200–203, New York, NY, USA. ACM.

Gutierrez, D. M. J., Solans, D., Heikkilä, M. A., Vitaletti, A., Kourtellis, N., Anagnostopoulos, A., and Chatzigiannakis, I. (2024). Non-iid data in federated learning: A survey with taxonomy, metrics, methods, frameworks and future directions. ArXiv, abs/2411.12377.

Huang, T., Lin, W., Shen, L., Li, K., and Zomaya, A. Y. (2022). Stochastic client selection for federated learning with volatile clients. IEEE Internet of Things Journal, 9(20):20055–20070.

Lobato, W., Costa, J. B. D. D., Souza, A. M. d., Rosário, D., Sommer, C., and Villas, L. A. (2022). Flexe: Investigating federated learning in connected autonomous vehicle simulations. In 2022 IEEE 96th Vehicular Technology Conference (VTC2022-Fall), pages 1–5.

Lobato, W., Da Costa, J. B., Gonzalez, L. F., Cerqueira, E., Rosário, D., Sommer, C., and Villas, L. A. (2024). Entropy and mobility-based model assignment for multi-model vehicular federated learning. In 2nd International Conference on Federated Learning Technologies and Applications (FLTA), pages 8–15. IEEE.

Mangipudi, S. et al. (2025). Cohere - congestion-aware offloading and handover via empirical rat evaluation for multi-rat networks. arXiv preprint arXiv:2511.00439.

Maroua, D. (2024). A state-of-the-art on federated learning for vehicular communications. Vehicular Communications, 45:100709.

Pan, H., Luo, M., Wang, J., Huang, T., and Sun, W. (2024). A Safe Motion Planning and Reliable Control Framework for Autonomous Vehicles. IEEE Transactions on Intelligent Vehicles, 9(4):4780–4793.

Smestad, C. and Li, J. (2023). A systematic literature review on client selection in federated learning. In Proceedings of the 27th International Conference on Evaluation and Assessment in Software Engineering, page 2–11. ACM.

Sousa, J., Ribeiro, E., Bustincio, R., Bastos, L., Morais, R., Cerqueira, E., and Rosário, D. (2025). Enhancing robustness in federated learning using minimal repair and dynamic adaptation in a scenario with client failures. Annals of Telecommunications, 80(9-10):885–899.

Vardhan, H., Yu, X., Rosing, T., and Mazumdar, A. (2025). Client Selection in Federated Learning with Data Heterogeneity and Network Latencies.

Zhang, B., Yu, W., Jia, Y., Huang, J., Yang, D., and Zhong, Z. (2023). Predicting vehicle trajectory via combination of model-based and data-driven methods using kalman filter. Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering, 238(8):2437–2450.

Zhang, C., Xie, Y., Bai, H., Yu, B., Li, W., and Gao, Y. (2021). A survey on federated learning. Knowledge-Based Systems, 216:106775.

Zhu, H., Xu, J., Liu, S., and Jin, Y. (2021). Federated learning on non-IID data: A survey. Neurocomputing, 465:371–390.