Optimal resource management and allocation for autonomous-vehicle-infrastructure cooperation under mobile edge computing

Purpose With the continuous technological development of automated driving and expansion of its application scope, the types of on-board equipment continue to be enriched and the computing capabilities of on-board equipment continue to increase and corresponding applications become more diverse. As the applications need to run on on-board equipment, the requirements for the computing capabilities of on-board equipment become higher. Mobile edge computing is one of the effective methods to solve practical application problems in automated driving. Design/methodology/approach In this study, in accordance with practical requirements, this paper proposed an optimal resource management allocation method of autonomous-vehicle-infrastructure cooperation in a mobile edge computing environment and conducted an experiment in practical application. Findings The design of the road-side unit module and its corresponding real-time operating system task coordination in edge computing are proposed in the study, as well as the method for edge computing load integration and heterogeneous computing. Then, the real-time scheduling of highly concurrent computation tasks, adaptive computation task migration method and edge server collaborative resource allocation method is proposed. Test results indicate that the method proposed in this study can greatly reduce the task computing delay, and the power consumption generally increases with the increase of task size and task complexity. Originality/value The results showed that the proposed method can achieve lower power consumption and lower computational overhead while ensuring the quality of service for users, indicating a great application prospect of the method.

Automated summary

This study proposed an optimal resource management allocation method of autonomous-vehicle-infrastructure cooperation in a mobile edge computing environment, which can greatly reduce task computing delay, with power consumption generally increasing with the increase of task size and complexity.