Predictive allocation of edge computing resources for autonomous driving

The project targets design and experimental verification of solutions for joint allocation of computing and mobile communication network resources for distributed processing of the complex applications composed of mutually dependent sub-tasks including those based on
neural networks. To guarantee reliability and availability of the resources on the time when these are required, we also consider pre-allocation (booking) of communication and computing resources in mid-term time horizon (dozens of second to hours) using blockchain.

Vehicular comunications, Edge computing, Radio resource management, Machine learning

Major results and publications

E. Slapak, A. Petik, M. Volosin, M. Dopiriak, J. Gazda and Z. Becvar
Graph Neural Network Empowered Resource Allocation for Connected Autonomous Mobility
IEEE Robotic Computing (IEEE IRC 2023), 2023.
M. Volosin, E. Slapak, Z. Becvar, T. Maksymyuk, A. Petik, M. Liyanage and , et al.
Blockchain-based Route Selection with Allocation of Radio and Computing Resources for Connected Autonomous Vehicles
IEEE Transactions on Intelligent Transportation Systems, volume 24, no. 7, 2023.

Date:: 07/2022 - 06/2025
Project No.:: LUASK22064
Funding:: Ministry of Education, Youth and Sports
Budget:: ~125k EUR
PI:: Zdenek Becvar
Team (6Gm):: Pavel Mach, Shahzeb Javed
Partners:: TU Kosice (prof. Juraj Gazda)
Topics:: Communications, computation, resource allocation, prediction, distributed machine learning, autonomous driving

-Ongoing-5G 6G MEC ML/AI Testbed VehicularPredictive allocation of edge computing resources for autonomous driving

Leave a Reply Cancel Reply