COGITO “A COGnItive dynamic sysTem to allOw buildings to learn and adapt” project is funded by the Italian Ministry of Education, University and Research (MIUR).(ARS01_00836)
It aims at integrating the IoT (Internet of Things) with cognitive dynamic systems (SDC) for an improved management of the public and residential buildings. The purpose is to lead an urban regeneration by redefining the living environments through innovative technological solutions which are able to improve the people’s quality of life.
The COGITO project was conceived to go beyond the concept of Smart Building Automation Systems (SBAS), overcoming the traditional model of building management based on ”a posteriori” analysis and actions, or with the analysis of historical data linked for example to the management of resources such as the energy use, water consumption, space occupancy rate, and then take action.
COGITO is novel because it uses the new fog/mist computing paradigm to create a cognitive embedded IoT system. In this approach, cognitive objects (sensors, actuators, controllers, cameras, etc.) can communicate with each other and with the Cloud. Moreover, they have processors on board to perform locally cognitive algorithms according to an analytics approach on the edge. All of this permits to locally analyze data, extract information that can be translated immediately into actions or recommendations, or can be passed to a higher conceptual level that offers a computational layer where a decision is made by evaluating global information.
With the use of cognitive objects, it is no longer necessary to use rule-based algorithms, but all the complexity of the management of the information is transferred to cognitive algorithms that will process information according to a data-driven approach. In such a case, users are only engaged in defining the characteristics of the raw data that they think may have relevant information, then, the functionalities and the actions to be taken are automatically learned. Automated learning systems, based on a data-driven approach, are excellent for analyzing large data like those produced by a large number of sensors.
The use of the cognitive IoT introduces a revolution in building automation systems, where decentralized architecture allows data analysis to be made on the edges of the network. This approach provides real-time intelligence and responsiveness in control operations. At the same time, it is possible to reduce bandwidth utilization as heavy communications are avoided. Current computing technology actually provides low-cost and high energy-save embedded processors that can be used for data analysis within the sensor itself. With this approach, data sent to the network can simply be the final result of a carried out analysis, which will have a smaller volume and will allow shorter response times.
The expected results of the project will help in developing innovative market-based cognitive services that can ensure building management through a unifying platform by applying data-driven technology to the building automation ecosystem. Proposed solutions will drive towards the future of building automation systems. A key feature is the possibility of implicitly taking into account occupants needs and preferences through a ”cognitive object model” which is able also to exploit the advantages related to the use of IoT devices equipped with computational capabilities allowing to support powerful and effective on-the-fly analysis. Newmarket services will make the buildings able to analyze space utilization, improving occupants comfort, and automatically portray intelligent management of the whole system. Cognitive Objects are complex objects that will benefit from innovative hardware technologies and advanced cognitive software algorithms. New embedded devices and/or dedicated processors (for example, Intel Movid- ius, NVIDIA Jetson Nano, Raspberry PI 3) will be evaluated to support the execution of cognitive algorithms. The use of such devices will allow the exploitation of a ”local” computational power that will permit the processing of data close to data sources in order to create a low-latency fog/mist computing environment. The buildings, by incorporating cognitive technologies, will increase the value of the offered services by making them more effective, cheaper, safer, faster and more distinctive. The new 5G communication technology, which will be tested for the realization of a demonstrator in the city of Matera, will bring significant performance improvements to support cognitive services in buildings.
The general objective of the project is to propose an open, free and accessible, interoperable and replicable “cognitive framework”, which creates an intelligent infrastructure for the supervision and control of all devices and objects in a building, taking into account the presence and behavior of people and implementing cognitive actions to optimize energy consumption, improve the safety of people and structures and encourage the active aging of people, sharing data and decisions according to a principle of active and inclusive involvement of citizens.
The framework through the presence of cognitive and social objects located in the building will make it possible to learn the behavior and habits of each user in order to anticipate their needs and support them in daily activities through personalized decisions.
The COGITO project is composed by 7 development objectives (ORs), 5 of which are related to Industrial Research and 2 are related to Experimental Development:
OR1 – Development of framework based on cognitive IoT for the efficient management of buildings
OR2 – Intelligent management of the building-plant system for energy efficiency and indoor quality
OR3 – Efficient management of building indoor and outdoor environments
OR4 – Audio/video edge analytics for management of security and control access
OR5 – Building physical management and damage restoration
OR6 – Experimentation in real sites: Realization of the Demonstrator – City of Rende (CS)
OR7 – Experimentation in real sites: Realization of the Demonstrator – City of Matera
Università della Calabria – UNICAL (Lead Partner); Consorzio SI-LAB; Attuatore: Consiglio Nazionale delle Ricerche – ICAR-CNR; Attuatore: OMNIA ENERGIA SpA 3. Agenzia nazionale per le nuove tecnologie, l’energia e lo sviluppo economico sostenibile – ENEA; Università degli Studi Mediterranea di Reggio Calabria – UNIRC; Università degli Studi della Basilicata – UNIBAS; Consorzio Universitario di Economia Industriale e Manageriale – CUEIM; Azienda Territoriale per l’Edilizia Residenziale Pubblica della Calabria – ATERP Calabria; SCAI Lab srl; Telecom Italia SpA; SITE SpA; Digimat srl; DS Tech srl; SMARTS srl
Funding Scheme MIUR – PON Ricerca e Innovazione 2014-2020 e FSC, Azione II. 2 Cluster Tecnologic
Total Grant: 8.585.926 Euro
ICAR Grant: 1.100.000 Euro
Application Area IoT & Cyber-Physical Systems
Start Date: 1 settembre 2018
End Date: 31 ottobre 2021
Project responsible: Giandomenico Spezzano, ICAR-CNR