Research and Analysis of Deep Learning for Distributed Autonomous Vehicles

Nanting Chen

doi:10.54097/q6v46n08

Authors

Nanting Chen School of Computer Science, Zhuhai College of Science and Technology, Zhuhai, 519000, China

DOI:

https://doi.org/10.54097/q6v46n08

Keywords:

Autonomous vehicles, Deep learning, Internet of Vehicles, Distributed computing, Multi-sensor fusion.

Abstract

Autonomous driving technology is evolving from individual vehicle intelligence to collective intelligence, and the deep integration of vehicle networking, distributed computing, and deep learning has become the key path to overcoming the bottlenecks of perception and computing power. This paper systematically reviews the three-stage development process of distributed computing, from cloud centralization and edge collaboration to intelligent collaboration, and constructs a technical framework encompassing data coordination, computing power scheduling, task allocation, and consistency assurance. It summarizes typical applications in the fields of environmental perception, vehicle status monitoring, communication optimization, collaborative decision-making, and security protection, and provides an in-depth analysis of the core challenges currently faced, including heterogeneous data integration, real-time constraints, communication reliability, security and privacy, and model generalization. Looking ahead, breakthroughs in technologies such as lightweight collaborative learning, physical knowledge embedding, privacy computing, integrated communication-computation-control, and digital twin will further deepen the integration of the three, laying a foundation for fully autonomous intelligent transportation systems.

References

[1] Hossain, N.M., Rahman, M.M., Tusher, H.E., et al. (2026) Optimised decision-making model for vehicle health monitoring system leveraging deep learning algorithm. Advanced Engineering Informatics,69(PA):103896-103896.

[2] Zhou, S., Yang, H., Lashkov, I., et al. (2025) Deep learning-based vehicle detection and tracking from roadside LiDAR data through robust affinity fusion. Expert Systems With Applications,279127338-127338.

[3] Onsu, A.M., Simsek, M., Fobert, M., et al. (2025) Intelligent multi-sensor fusion and anomaly detection in vehicles via deep learning. Internet of Things,31101561-101561.

[4] Liu, J., Li, Y., Dong, S., et al. (2025) Overloaded vehicle identification for long-span bridges based on physics-informed multi-task deep learning leveraging influence line. Engineering Structures,333120163-120163.

[5] Li, X., Ren, M. (2025) Road adhesion coefficient Estimation: Physics-informed deep learning method with vehicle dynamics model. Expert Systems With Applications,260125387-125387.

[6] Yadav, L. A., Goyal, K.S. (2024) An Efficient and Intelligent System for Controlling the Speed of Vehicle using Fuzzy Logic and Deep Learning. International Journal of Advanced Computer Science and Applications (IJACSA),15(3).

[7] Aswal, K., Pathak, H. (2024) Advancing Vehicle Security: Deep Learning based Solution for Defending CAN Networks in the Internet of Vehicles. EAI Endorsed Transactions on Internet of Things,10(1).

[8] Osman, A.R. (2023) Optimizing Autonomous Vehicle Communication through an Adaptive Vehicle-to-Everything (AV2X) Model: A Distributed Deep Learning Approach. Electronics,12(19).

[9] Ragab, M., Abdushkour, A.H., Khadidos, O.A., et al. (2023) Improved Deep Learning-Based Vehicle Detection for Urban Applications Using Remote Sensing Imagery. Remote Sensing,15(19).

[10] Ahmed, R.O., Nagy, S.S., et al. (2021) Enhancing the Reliability of Communication between Vehicle and Everything (V2X) Based on Deep Learning for Providing Efficient Road Traffic Information. Applied Sciences,11(23):11382-11382.

[11] Hassam, T., EunSung, J. (2023) Comparative Study on Distributed Lightweight Deep Learning Models for Road Pothole Detection. Sensors (Basel, Switzerland),23 (9).