The AI@EDGE Consortium has released this summer its first technical Deliverable: D2.1 “Use cases, requirements, and preliminary system architecture” as main outcome of the Task 2.1. This report details the activity performed to characterize and specify the four AI@EDGE use cases – UC1 / Platform for virtual validation of vehicle cooperative perception; UC2 / Platform for secure and resilient orchestration of large (I)IoT networks; UC3 / Platform for edge AI assisted drones in beyond-visual-light-of-sight operations; UC4 / Platform of smart content & data curation for in-flight entertainment services.
The deliverable caters preliminary architectural and system requirements for AI@EDGE; in particular, the type of users, services, and applications defined by the use cases, in order to derive application driven network automation requirements. Furthermore, system-level and functional requirements are defined, considering the potentials of the serverless paradigm combined with AI-enabled network operations.
Task 2.1 effort is meant to contribute to the definition of a set of architectural recommendations for the 5G-PPP program with particular focus on the realization of sustainable beyond-5G and 6G architectures capable of supporting the innovative services and use cases envisioned by AI@EDGE.
This deliverable draws a preliminary functional view on the 5G AI@EDGE system architecture that will be further developed in D2.2 “Preliminary assessment of system architecture, interfaces specifications, and techno-economic analysis”; D3.1 “Initial report on systems and methods for AI@EDGE platform automation” and D4.1 “Design and initial prototype of the AI@EDGE connect-compute platform”, all due in December 2021.
If you have missed the release of D2.1, you can read it here.