Research Article
A Fog-Based IOV for Distributed Learning in Autonomous Vehicles
@INPROCEEDINGS{10.1007/978-3-030-94763-7_18, author={Pawan Subedi and Beichen Yang and Xiaoyan Hong}, title={A Fog-Based IOV for Distributed Learning in Autonomous Vehicles}, proceedings={Mobile Networks and Management. 11th EAI International Conference, MONAMI 2021, Virtual Event, October 27-29, 2021, Proceedings}, proceedings_a={MONAMI}, year={2022}, month={1}, keywords={Internet of Vehicles Fog computing Connected vehicles Federated learning Autonomous driving}, doi={10.1007/978-3-030-94763-7_18} }- Pawan Subedi
Beichen Yang
Xiaoyan Hong
Year: 2022
A Fog-Based IOV for Distributed Learning in Autonomous Vehicles
MONAMI
Springer
DOI: 10.1007/978-3-030-94763-7_18
Abstract
Internet of Vehicles (IoVs) consist of connected vehicles and connected autonomous vehicles. With fog computing built within the IoV, it becomes promising for federated learning to be used in vehicular environments. One important application of such a fog computing system is distributed deep learning for decision-making tasks in autonomous driving. In this paper, a distributed training system building on top of the Named-Data Networking (NDN) architecture is introduced in order to combat the mobility challenges to the underlying network. The paper presents analyses on critical latency issues pertained to soliciting the worker CVs and collecting the partial updates. Further, the advantages of using NDN for the IoV are evaluated with comparisons to IP network through simulation. The results show promising performance gains for the evaluation cases.
