Self-Driving Car Meets Multi-Access Edge Computing for Deep Learning-Based Caching

Engineering notes

Key topics: self-driving car, self-driving, end-to-end, prediction. See the paper for implementation details and experimental results.

Chinese explanation / 中文解读

中文解读待补充：本站会优先为端到端自动驾驶、BEV感知、3D目标检测、轨迹预测、路径规划、LiDAR感知等高价值论文补充中文说明。

Original abstract

In the future, self-driving cars are expected to be involved in public transportation. Once passengers are comfortable with them, the self-driving cars will be new spaces for entertainment. However, getting infotainment contents from Data Centers (DCs) can be perturbed by the high end-to-end delay. To address this issue, we propose caching for infotainment contents in close proximity to the self-driving cars and in self-driving cars. In our proposal, Multi-access Edge Computing (MEC) helps self-driving cars by deploying MEC servers to the edge of the network at macro base stations (BSs), WiFi access points (WAPs), and roadside units (RSUs) for caching infotainment contents in close proximity to the self-driving cars. Based on the passenger's features learned via self-driving car deep learning approach proposed in this paper, the self-driving car can download infotainment contents that are appropriate to its passengers from MEC servers and cache them. The simulation results show that our prediction for the infotainment contents need to be cached in close proximity to the self-driving cars can achieve 99.28% accuracy.

Links and sources

• Official / arXiv page

Comments