Low-Latency Edge Learning Framework for Real-Time Decision-Making in Autonomous Driving
DOI:
https://doi.org/10.71222/af3x9q45Keywords:
edge computing, federated learning, real-time decision-making, autonomous driving control, distributed model trainingAbstract
The study introduces a real-time decision-making framework tailored for autonomous driving environments, aiming to address the critical challenges of latency reduction, data privacy, and decentralized learning. The proposed architecture leverages edge computing infrastructure in combination with federated learning to enable collaborative model training across vehicle-end and roadside units. A Federated Averaging (FedAvg) strategy, augmented with differential privacy techniques, is employed to safeguard sensitive information and enhance model stability. To minimize communication costs and computational overhead at the edge, the framework integrates sparse update mechanisms and model compression via pruning. The effectiveness of the proposed system is verified through extensive experiments conducted on the CARLA simulation platform and in real-world deployment scenarios. Results indicate a 31.2% decrease in decision-making latency, while maintaining on-device data training. Additionally, the framework demonstrates improved path planning accuracy and adaptability under dynamic, interactive traffic conditions.
References
1. D. Garikapati and S. S. Shetiya, "Autonomous vehicles: Evolution of artificial intelligence and the current industry landscape," Big Data Cogn. Comput., vol. 8, no. 4, p. 42, 2024, doi: 10.3390/bdcc8040042.
2. Y. Yao, "Design of neural network-based smart city security monitoring system," in Proc. 2024 Int. Conf. Comput. Multimedia Technol., pp. 275-279, May 2024, doi: 10.1145/3675249.3675297.
3. J. Monios and R. Bergqvist, "The transport geography of electric and autonomous vehicles in road freight networks," J. Transp. Geogr., vol. 80, p. 102500, 2019, doi: 10.1016/j.jtrangeo.2019.102500.
4. Q. Yu, S. Wang, and Y. Tao, "Enhancing anti-money laundering detection with self-attention graph neural networks," SHS Web Conf., vol. 213, p. 01016, 2025, doi: 10.1051/shsconf/202521301016.
5. Y. Yan, Y. Wang, J. Li, J. Zhang, and X. Mo, "Crop yield time-series data prediction based on multiple hybrid machine learning models," Comput. Electron. Agric., 2025, doi: 10.48550/arXiv.2502.10405.
6. S. B. Chougule, B. S. Chaudhari, S. N. Ghorpade, and M. Zennaro, "Exploring computing paradigms for electric vehicles: From cloud to edge intelligence, challenges and future directions," World Electr. Veh. J., vol. 15, no. 2, p. 39, 2024, doi: 10.3390/wevj15020039.
7. Y. Xiao, L. Tan, and J. Liu, "Application of machine learning model in fraud identification: A comparative study of CatBoost, XGBoost and LightGBM," Comput. Ind. Eng., 2025, doi:10.20944/preprints202503.1199.v1
8. J. Wang, W. Ding, and X. Zhu, "Financial analysis: Intelligent financial data analysis system based on LLM-RAG," IEEE Trans. Knowl. Data Eng., 2025, doi: 10.48550/arXiv.2504.06279.
9. T. Kujala, J. Mäkelä, I. Kotilainen, and T. Tokkonen, "The attentional demand of automobile driving revisited: Occlusion dis-tance as a function of task-relevant event density in realistic driving scenarios," Hum. Factors, vol. 58, no. 1, pp. 163-180, 2016, doi: 10.1177/0018720815595901.
10. A. Vepa, Z. Yang, A. Choi, J. Joo, F. Scalzo, and Y. Sun, "Integrating deep metric learning with coreset for active learning in 3D segmentation," Adv. Neural Inf. Process. Syst., vol. 37, pp. 71643-71671, 2024, doi: 10.48550/arXiv.2411.15763.
11. Z. Li, Q. Ji, X. Ling, and Q. Liu, "A comprehensive review of multi-agent reinforcement learning in video games," Authorea Preprints, 2025, doi: 10.36227/techrxiv.173603149.94954703/v1.
12. T. Gindele, S. Brechtel, and R. Dillmann, "Learning driver behavior models from traffic observations for decision making and planning," IEEE Intell. Transp. Syst. Mag., vol. 7, no. 1, pp. 69-79, 2015, doi: 10.1109/MITS.2014.2357038.
13. W. Zhang, Z. Li, and Y. Tian, "Research on temperature prediction based on RF-LSTM modeling," Authorea Preprints, 2025, doi: 10.36227/techrxiv.173603336.69370585/v1.
14. J. P. V. Talusan, M. Wilbur, A. Dubey, and K. Yasumoto, "Route planning through distributed computing by road side units," IEEE Access, vol. 8, pp. 176134-176148, 2020, doi: 10.1109/ACCESS.2020.3026677.
15. J. Liu, et al., "Application of deep learning-based natural language processing in multilingual sentiment analysis," Mediterr. J. Basic Appl. Sci., vol. 8, no. 2, pp. 243-260, 2024, doi: 10.46382/MJBAS.2024.8219.
16. X. Tang, Z. Wang, X. Cai, H. Su, and C. Wei, "Research on heterogeneous computation resource allocation based on data-driven method," in Proc. 6th Int. Conf. Data-driven Optim. Complex Syst. (DOCS), 2024, pp. 916-919, doi: 10.1109/DOCS63458.2024.10704406.
17. H. Feng, "The research on machine-vision-based EMI source localization technology for DC-DC converter circuit boards," in Proc. 6th Int. Conf. Inf. Sci., Electr. Autom. Eng. (ISEAE 2024), vol. 13275, pp. 250-255, Sep. 2024, doi: 10.1117/12.3037693.
18. J. Zhu, J. Ortiz, and Y. Sun, "Decoupled deep reinforcement learning with sensor fusion and imitation learning for autonomous driving optimization," in Proc. 6th Int. Conf. Artif. Intell. Comput. Appl. (ICAICA), 2024, pp. 306-310, doi: 10.1109/ICAICA63239.2024.10823066.
19. H. G. Abreha, M. Hayajneh, and M. A. Serhani, "Federated learning in edge computing: A systematic survey," Sensors, vol. 22, no. 2, p. 450, 2022, doi: 10.3390/s22020450.
20. J. Zhu, Y. Sun, Y. Zhang, J. Ortiz, and Z. Fan, "High fidelity simulation framework for autonomous driving with augmented reality based sensory behavioral modeling," in Proc. IET Conf. Proc. CP989, vol. 2024, no. 21, pp. 670-674, Oct. 2024, doi: 10.1049/icp.2024.4298.
21. S. Goyal, A. Griggio, and S. Tonetta, "System-level simulation-based verification of Autonomous Driving Systems with the VIVAS framework and CARLA simulator," Sci. Comput. Program., vol. 242, p. 103253, 2025, doi: 10.1016/j.scico.2024.103253.
22. Y. Sun, N. S. Pargoo, P. J. Jin, and J. Ortiz, "Optimizing autonomous driving for safety: A human-centric approach with LLM-enhanced RLHF," in Companion of the 2024 ACM Int. Joint Conf. Pervas. Ubiquitous Comput., pp. 76-80, Oct. 2024, doi: 10.1145/3675094.3677588.
23. F. Qin, H. Y. Cheng, R. Sneeringer, M. Vlachostergiou, S. Acharya, H. Liu, and L. Yao, "ExoForm: Shape memory and self-fusing semi-rigid wearables," in Extended Abstr. 2021 CHI Conf. Human Factors Comput. Syst., pp. 1-8, May 2021, doi: 10.1145/3411763.3451818.
24. H. Iqbal, D. Campo, P. Marin-Plaza, L. Marcenaro, D. M. Gómez, and C. Regazzoni, "Modeling perception in autonomous vehicles via 3D convolutional representations on LiDAR," IEEE Trans. Intell. Transp. Syst., vol. 23, no. 9, pp. 14608-14619, 2021, doi: 10.1109/TITS.2021.3130974.
25. K. Mo, et al., "Dral: Deep reinforcement adaptive learning for multi-UAVs navigation in unknown indoor environment," arXiv preprint arXiv:2409.03930, 2024, doi: 10.48550/arXiv.2409.03930.
26. T. S. Darwish and K. A. Bakar, "Fog-based intelligent transportation big data analytics in the internet of vehicles environment: Motivations, architecture, challenges, and critical issues," IEEE Access, vol. 6, pp. 15679-15701, 2018, doi: 10.1109/ACCESS.2018.2815989.
27. X. Shi, Y. Tao, and S. C. Lin, "Deep neural network-based prediction of B-cell epitopes for SARS-CoV and SARS-CoV-2: En-hancing vaccine design through machine learning," in Proc. 4th Int. Signal Process., Commun. Eng. Manag. Conf. (ISPCEM), pp. 259-263, Nov. 2024, doi: 10.1109/ISPCEM64498.2024.00050.
28. S. Park, D. Kim, and S. Lee, "Enhancing V2X security through combined rule-based and DL-based local misbehavior detection in roadside units," IEEE Open J. Intell. Transp. Syst., 2024, doi: 10.1109/OJITS.2024.3479716.
29. J. Zhu, Y. Wu, Z. Liu, and C. Costa, "Sustainable optimization in supply chain management using machine learning," Int. J. Manag. Sci. Res., vol. 8, no. 1, 2025, doi: 10.53469/ijomsr.2025.08(01).01.
30. S. Wang, R. Jiang, Z. Wang, and Y. Zhou, "Deep learning-based anomaly detection and log analysis for computer networks," 2024, arXiv preprint arXiv:2407.05639, doi: 10.48550/arXiv.2407.05639.
31. C. Gong, X. Zhang, Y. Lin, H. Lu, P. C. Su, and J. Zhang, "Federated learning for heterogeneous data integration and privacy protection," Comput. Sci. Eng., 2025, doi: 10.20944/preprints202503.2211.v1.
32. D. Luo, J. Gu, F. Qin, G. Wang, and L. Yao, "E-seed: Shape-changing interfaces that self-drill," in Proc. 33rd Annu. ACM Symp. User Interface Softw. Technol., pp. 45-57, Oct. 2020, doi: 10.1145/3379337.3415855.
33. J. Gu, et al., "Inverse design tool for asymmetrical self-rising surfaces with color texture," in Proc. 5th Annu. ACM Symp. Comput. Fabrication, pp. 1-12, Nov. 2020, doi: 10.1145/3424630.3425420.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Xiangwei Liu, Meilin Zhang, Ka-Ho Chan, Wing-Yee Leung, Zhihao Wu (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
