EMPYREAN envisages a hyper-distributed computing paradigm, based on federations of collaborative and heterogeneous IoT devices and resources (e.g., on RISC-V) across different providers and networks. These federations, namely Associations, operate autonomously and interconnect seamlessly utilizing distributed, cognitive and dynamic AI-enabled decision-making, to balance computing tasks and data inside an Association as well as between Associations in a multi-agent manner and across central computing environments, optimizing resources and providing scalability, resiliency, energy efficiency and quality of service. An Association will constitute a trusted execution environment, while identity and data access management schemes will assure controlled access and confidentiality of data, utilizing Cluster 3 related outcomes from participating partners. EMPYREAN will also be empowered with automated tools and mechanisms for efficient data processing of AI-workloads and secure distributed edge storage. Developed technologies will also enable Associations-native application development and deployment, contributing to the entire application lifecycle and interoperability. EMPYREAN will provide open and standardized APIs, while utilizing and extending open-source platforms maintained by European companies from the consortium. EMPYREAN will demonstrate its advanced and innovative capabilities through three well-defined use cases that involve device- and data-rich applications in advanced manufacturing, smart agriculture and warehouse automation, involving AI-driven value extraction from high volume and dynamic IoT data generated by multiple sources (e.g., robots) at the edge of the network. Also, a South Korea based use case in smart factories will further showcase the benefits of the EMPYREAN’s technologies. EMPYREAN will develop its Association-based continuum through synergies with emerging IPCEI initiatives and EU bodies (e.g., GAIA-X, IDSA) by involved partners.

Over the past decades the mobile communications has evolved over the different generations to the current 5G, and transformed into a fundamental infrastructure that supports digital demands from all industry sectors. However, 5G systems are expected to fall short on meeting the anticipated stringent performance requirements for the new generation of real time mission-critical applications. In view of that, DESIRE6G will design and develop novel zero-touch control, management, and orchestration platform, with native integration of AI, to support eXtreme URLLC application requirements. DESIRE6G will re-architect mobile networks through a) its intent-based control and end-to-end orchestration that targets to achieve near real time autonomic networking; and b) a cloud-native unified programmable data plane layer supporting multi-tenancy. The latter will be supported by a generic hardware abstraction layer designed for heterogeneous systems. Flexible composition of modular micro-services for slice specific implementations and flexible function placement depending on HW requirements will enable granular use case instantiation and service level assurance with minimum resource consumption and maximum energy efficiency. The DESIRE6G data, control, management, and orchestration plane is supported by a pervasive monitoring system, extending from the network to the user equipment or IoT terminal. DESIRE6G will employ distributed ledger technology to support a) dynamic federation for services across of multiple administrative domains and b) infrastructure-agnostic software security. Finally, DESIRE6G will enable communication-, and energy- efficient distributed AI, at the network edge, while considering application-level requirements and resource constraints. The proposed innovations will be validated employing a VR/AR/MR and a Digital Twin application at two distinct experimental sites.

TEXTaiLES goal is to redefine the digitization of Cultural Heritage (CH) textile objects and the development of AI-based processing tools, aiming to establish a standardized protocol that seamlessly integrates into the ECCCH. By studying past and present practices in textile digitization, we methodically explore their strengths and limitations. This comprehensive assessment, enriched by contributions from co-designer groups across various use cases, helps us identify the essentials of textile digitization. Our approach to standardization is all-encompassing. We develop advanced tools that address the entire digitization life cycle of textile objects, focusing on non-destructive methods. Using robotic, cost-effective multi-sensor solutions, we ensure the preservation of artefacts during digitization, complemented by reliable data analysis workflows. Innovation lies within the core of TEXTaiLES. We integrate state-of-the-art AI technologies to shed light on the intricate details of textiles. From revealing their weaving patterns and microstructures to identifying degradation factors, our AI-driven toolkit offers in-depth insights. It further excels in motion modeling, dynamics, and the reconstruction of fragmented or lost objects. All these capabilities merge into a groundbreaking 4D multi-scale annotation interface, coupled with a digital twin model and enhanced by human-in-loop interactions. Our commitment to the CH community remains strong. We aim to embed the tools and techniques developed by TEXTaiLES into archaeological research, facilitated by the ECCCH. By supporting interdisciplinary research communities and providing upskilling opportunities, we aim to raise the textile CH domain to new heights.

MLSysOps will achieve substantial research contributions in the realm of AI-based system adaptation across the cloud-edge continuum by introducing advanced methods and tools to enable optimal system management and application deployment. MLSysOps will design, implement and evaluate a complete framework for autonomic end-to-end system management across the full cloud-edge continuum. MLSysOps will employ a hierarchical agent-based AI architecture to interface with the underlying resource management and application deployment/orchestration mechanisms of the continuum. Adaptivity will be achieved through continual ML model learning in conjunction with intelligent retraining concurrently to application execution, while openness and extensibility will be supported through explainable ML methods and an API for pluggable ML models. Flexible/efficient application execution on heterogeneous infrastructures and nodes will be enabled through innovative portable container-based technology. Energy efficiency, performance, low latency, efficient, resilient and trusted tier-less storage, cross-layer orchestration including resource-constrained devices, resilience to imperfections of physical networks, trust and security, are key elements of MLSysOps addressed using ML models. The framework architecture disassociates management from control and seamlessly interfaces with popular control frameworks for different layers of the continuum. The framework will be evaluated using research testbeds as well as two real-world application-specific testbeds in the domain of smart cities and smart agriculture, which will also be used to collect the system-level data necessary to train and validate the ML models, while realistic system simulators will be used to conduct scale-out experiments. The MLSysOps consortium is a balanced blend of academic/research and industry/SME partners, bringing together the necessary scientific and technological skills to ensure successful implementation and impact.