GAIA-X 4 AGEDA: A Federated and Secure Cloud Framework with AI-Driven Traffic Compliance in Connected Vehicles

• Achim Rettberg Hamm-Lippstadt University of Applied Sciences
• Narmada Ambigapathy Hamm-Lippstadt University of Applied Sciences
• Fatima Idrees Hamm-Lippstadt University of Applied Sciences
• Ali Alhalabi Hamm-Lippstadt University of Applied Sciences
• Katrin Glöwing Hamm-Lippstadt University of Applied Sciences
• Mehdi Azarafza Hamm-Lippstadt University of Applied Sciences

Resumo

Connected and autonomous vehicles increasingly rely on secure cloud infrastructures to exchange perception, safety, and compliance information. Yet, the lack of standardized mechanisms for trustworthy data federation limits interoperability and regulatory validation across heterogeneous vehicle fleets. This paper presents GAIA-X 4 AGEDA, a federated and secure cloud framework that combines dynamic edge orchestration with AI-driven traffic compliance monitoring for connected vehicles. The architecture leverages the GAIA-X Trust Framework for credential-based authentication and sovereign data exchange, while integrating Orchestration for mixed-criticality workload management across a four-zone vehicle architecture. An edge-AI compliance module validates real-time adherence to legal road traffic regulations, with results securely transmitted through GAIA-X Data Spaces to enable cooperative awareness among connected vehicles. Using a CARLA-based proof-of-concept [1] deployed on Raspberry Pi 5 and Jetson Nano P3541 platforms, we demonstrate how compliance violations detected at the edge are authenticated, federated, and shared with Traffic Awareness Services (TAS) while preserving data sovereignty. The framework achieves V2X compliance alerts and this work establishes a practical pathway toward dependable, regulation-aware, and federated vehicle ecosystems aligned with European data sovereignty principles, enabling secure service deployment throughout the vehicle lifecycle.

Palavras-chave: GAIA-X 4 AGEDA, federated data spaces, connected vehicles, traffic compliance, edge orchestration, AI-driven monitoring, cloud federation, software-defined vehicles

Referências

A. Dosovitskiy, G. Ros, F. Codevilla, A. Lopez, and V. Koltun, CARLA: An Open Urban Driving Simulator, arXiv preprint arXiv:1711.03938, 2017. [Online]. Available: [link]

GAIA-X 4 Future Mobility, “GAIA-X 4 AGEDA Project,” Accessed: Oct. 10, 2025. [Online]. Available: [link]

GAIA-X AISBL, “GAIA-X Framework,” Accessed: Oct. 10, 2025. [Online]. Available: [link]

N. Ambigapathy, C. Steinmetz, L. Stahlbock, K. Rusev, A. Kos, S. A. Böttigheimer, and A. Rettberg, “GAIA-X 4 AGEDA Powered Software-Defined Vehicles: Hybrid Failure-Aware Orchestration for Adaptive Mixed-Criticality Workload Management,” in Proc. (IAVVC), pp. 1–6. [Online]. DOI: 10.1109/IAVVC61942.2025.11219638

S. Semerikov, T. Vakaliuk, O. Kanevska, M. Moiseienko, I. Donchev, and A. Kolhatin, “LLM on the edge: the new frontier,” in Proc. CEUR Workshop Vol. 3943, Mar. 2025.

M. A. Onsu, P. Lohan, B. Kantarci, A. Syed, M. Andrews, and S. Kennedy, “Semantic edge–cloud communication for real-time urban traffic surveillance with ViT and LLMs over mobile networks,” arXiv preprint arXiv:2509.21259, 2025. [Online]. Available: [link]

F. Idrees, N. Ambigapathy, P. Adelt, and A. Rettberg, “AI-Driven Traffic Compliance in Autonomous Vehicles: A Scalable Framework with Large Language Models on Edge Platforms,” in Proc. (IAVVC), pp. 1–6. [Online]. DOI: 10.1109/IAVVC61942.2025.11219567

GAIA-X - “MoveID Project,” Accessed: Oct. 10, 2025. [Online]. Available: [link]

German Road Traffic Regulations (StVO), Bundesministerium für Digitales und Verkehr.

A. Radford, J. Wu, R. Child, D. Luan, D. Amodei, and I. Sutskever, “Language models are unsupervised multitask learners,” in Proceedings, 2019.

L. Gao, S. Biderman, S. Black, L. Golding, T. Hoppe, C. Foster, J. Phang, H. He, A. Thite, N. Nabeshima, S. Presser, and C. Leahy, “The Pile: An 800GB dataset of diverse text for language modeling,” CoRR, vol. abs/2101.00027, 2021.

S. Zhang, S. Roller, N. Goyal, M. Artetxe, M. Chen, S. Chen, C. Dewan, M. Diab, X. Li, X. V. Lin, T. Mihaylov, M. Ott, S. Shleifer, K. Shuster, D. Simig, P. S. Koura, A. Sridhar, T. Wang, and L. Zettlemoyer, “OPT: Open pre-trained transformer language models,” arXiv, 2022.

BigScience Workshop, “BLOOM: A 176B-parameter open-access multilingual language model,” arXiv e-prints, no. arXiv:2211.05100, 2022.

V. Sanh, L. Debut, J. Chaumond, and T. Wolf, “DistilGPT-2: a distilled, faster, and lighter version of GPT-2,” Hugging Face, 2019. [Online]. Available: [link]

H. W. Chung, L. Hou, S. Longpre, B. Zoph, Y. Tay, W. Fedus, E. Li, X. Wang, M. Dehghani, S. Brahma, A. Webson, S. Gu, Z. Dai, M. Suzugun, X. Chen, A. Chowdhery, J. Devlin, A. Roberts, D. Zhou, Q. Le, J. Wei, “Scaling Instruction-Finetuned Language Models,” arXiv preprint arXiv:2210.11416, 2022. [Online]. Available: [link]