ICAR, Institute for high performance computing and networking, is an Institute of the National Research Council (CNR) that belongs to the Department of Engineering, ICT and Technologies for Energy and Transport (DIITET). ICAR is present on the national territory with three offices, Rende, Napoli and Palermo. ICAR offers innovative solutions in terms of research, technology transfer and high education in the area of intelligent systems with complex functionality (cognitive systems and robotics, knowledge representation, extraction and reasoning, human-computer interface, optimization) and high performance computing (cloud computing, parallel and distributed environments and advanced technologies for the Internet). Significant applications are developed in the areas of e-health, energy, security, bioinformatics, cultural heritage, smart cities.
The Behavioral Modeling and Scalable Analytics (BMSA) group includes researchers with experience in the area of Behavior Computing (i.e. the definition of mathematical and computational models that can represent and forecast the behavior of entities operating in complex environments) and Cognitive Environments in Physical Buildings (i.e., effective methodologies and approaches for improving the management of public and residential buildings by enriching them with cognitive and self-adaptive capabilities).
The BMSA group exhibits significant experience in the adoption of Cognitive Objects (complex objects relying on innovative hardware and software solutions) and edge computing devices (dedicated sensors and actuation systems exploited to promote “local” computation), gained through projects initiatives within the context of the National Operational Programme for Research and Innovation, such as “Distretto Tecnologico DOMUS – Piattaforma intelligente per il monitoraggio e la gestione della sicurezza in-home di persone e strutture” and “COGITO - Sistema dinamico e cognitivo per consentire agli edifici di apprendere ed adattarsi”. Worthy of attention are the research activities carried out in the context of IoT platforms tailored for the development of general cognitive cyber–physical systems and suitable for the design and implementing smart city services and applications. Such platforms: (i) leverage the edge computing paradigm in a continuum with cloud computing through both the exploitation of the agent metaphor and a distributed network of computing nodes directly scattered in the physical environment, and (ii) enable the dynamic deployment of new computing nodes as well as software agents for addressing geographical and functional extensibility. A set of abstractions are offered for hiding the heterogeneity of the physical sensing/actuator devices embedded in the system, and for supporting in-network computation and
complex machine learning services. Effective services have been developed for improving end-user experiences, by managing thermal and visual comfort while optimizing energy consumption, and promoting the use of green energy and green-aware behavior.
In addition, the group has extensive experience in devising novel ML techniques that can be applied on a broad spectrum of data models including graphs, trees, unstructured data, and sensor/stream data also able to deal with the volume, variety and velocity of complex big data. Experience in this context was gained through the initiatives SecureOpenNets - “Distributed Ledgers For Secure Open Communities” (aimed at studying advanced solutions for privacy, sharing-economy and digital-rights, leveraging an innovative combination of Blockchain, Internet of Value/Things and Artificial Intelligence techniques), True Detective 4.0 (aimed at defining a platform, based on advanced analytics, that exploits sensor data within a smart manufacturing environment), Catch 4.0 (focused on providing solutions for improving content media management and advertising), and Humane-AI-Net (aiming at developing and maintaining a EU-wide network of excellence centers for a “Human-Centric” evolution of AI solutions).
