Don’t Let Me Down! Offloading Robot VFs Up to the Cloud

Robot functions are increasingly offloaded to the Edge to meet the strict latency needs of networked robotics, but Edge resources are expensive and the Cloud is sometimes a viable, cheaper option. The paper proposes Don’t Let Me Down! (DLMD), an algorithm that pushes robot virtual functions to the Cloud whenever possible to minimize Edge usage, while respecting the service latency constraints, wireless connectivity and robot mobility. DLMD also takes function migration and radio handover decisions to avoid excessive latency and loss of coverage.

Citation

K. Gillani, J. Martín-Pérez, M. Groshev, A. de la Oliva, and R. Gazda, “Don’t Let Me Down! Offloading Robot VFs Up to the Cloud,” in Proc. IEEE 9th International Conference on Network Softwarization (NetSoft), Madrid, Spain, 2023, pp. 232–236, doi: 10.1109/NetSoft57336.2023.10175499.

Authors from IE Research Datalab