Home » Gruppi Di Ricerca » 18328
Networked Systems

The Networked Systems research group, led by Prof. Antonio Pietrabissa, aims at developing control methodologies in the context of networked systems.

Besides classical control methods, such as model predictive control, optimal control and robust control, distributed non-cooperative control methods are being developed on the ground of mean-field game theory as well as learning methodologies such as reinforcement learning and deep reinforcement learning. Application areas of interest are communication networks, energy distribution networks, cyber-physical security in interconnected systems, bioengineering (e.g., brain connectivity).

The Networked Systems group members cooperate with researchers from national and international academia and industries. Among other collaborations, currently the group’s members are working with ETRI (Electronics and Telecommunications Research Institute), which is the most important research institute in South Korea, CEA (Commissariat à l'énergie atomique et aux énergies alternatives), which is the French research organisation in the areas of energy, security, information technologies and health technologies, the Faculty of Control Systems and Robotics, St. Petersburg, Russia, the Université libre de Bruxelles (ULB), Belgium, the University of Coimbra, Portugal, Tunghai University, Taiwan. See the list of the group's external members and of the group's publications for further information.

Currently, the main research topics of the group are the ones listed below.

  • Future Internet
    The group’s research supports a Future Internet vision, on the ground of the participation in the large FI-WARE EU-funded project concerning the Future Internet technology foundation and in projects on 5G communications, to develop a technology-independent distributed framework including coordinated control algorithms. These algorithms, based on homogeneous metadata describing the network and user status, manage the network resources and services to maximize the resource exploitation while satisfying the user requirements. The adopted methodologies include model-free learning, multi-agent systems, cross-layer/cross-network optimization, context awareness, data fusion.

     

  • e-Health
    The focus of the research activities of the group is related to the design and development of Intelligent Systems to support medical workers in the diagnosis and treatment processes. The group has studied several solutions for medical imaging analytics to provide medical operators with detailed reports of the anomalies and key features detected inside a decision support system. The group has developed customised algorithms for Federated Learning, to allow a GDPR-compliant knowledge sharing solution among networked clinical institutions by enabling the training of distributed Artificial Intelligence systems. Recent research activities focus also on the design of predictive and individualised control algorithms for the insulin treatment of patients using an artificial pancreas.

     

  • Smart Energy
    The research group tackles several control problems in the smart grid and power systems domain, including: control of renewable energy sources, active demand and demand side management in the residential and commercial sectors, algorithms for smart charging control of plug-in electric vehicles, integration of storage and other distributed energy resources into the grid. The research group has cooperated with several italian and european research centers, universities and industries in many national and european research projects, where it has developed smart grid control algorithms mostly based on model predictive control and nonlinear control techniques (e.g., feedback linearization).

     

  • Space
    Within this topic, the research group aims at developing control methodologies in the context of space-related applications, such as satellite communication networks along with their interaction with terrestrial (wired and wireless) ones, satellite networks used for emergency prevention, satellite launchers, sensor networks for planetary explorations. The control methodologies are applied in several international research projects funded by ESA and EU and range from classical feedback control of time-delay systems for congestion control problems to distribued non-cooperative control for load balancing and routing problems and deep reinforcement solutions for admission control problems.

The group members are involved in the activities of the Consortium for the Research in Automation and Telecommunications (CRAT), whose members are University of Rome Sapienza, Politecnico di Bari, University of Sannio, Thales Alenia Space Italia and TopNetwork. The aim of CRAT is to carry out applied research in the context of National and European projects and to favour the birth of start-ups. In 2013, the Sapienza start-up Ares2t was funded by members of the Networked System group on the ground of research in the field of smart grids and smart charging of electric vehicles.

On-going research projects

  • FedMedAI, Elaborazione di dati clinici con metodologie di intelligenza artificiale per strutture sanitarie federate nel rispetto del GDPR, April 2021-April 2023, Prot. n. A0375-2020-36491 del 23/10/2020, https://sites.google.com/diag.uniroma1.it/fedmedai/home
  • VADUS, Virtual Access and Digitalization for Unreachable Sites,  October 2020-October 2022, European Space Agency (ESA), 5G for L’ART programme.
  • ARIES, Advanced multi-Rat Integrated multi-sensors solution for Emergency prevention, detection and response operationS (managed by CRAT), November 2020-November 2021, European Space Agency (ESA), 5G for L’ART programme.
  • 5G-ALLSTAR, 5G AgiLe and fLexible integration of SaTellite And cellular (managed by CRAT), July 2018-June 2021, H2020-EUK2018.
  • 5G-SOLUTIONS, 5G Solutions for European Citizens (managed by CRAT), June 2019-May 2022, EU H2020-ICT-2019.
  • SESAME, Smart European Space Access thru Modern Exploitation of Data Science (managed by CRAT), January 2015 - December 2022, EU H2020-SPACE-16-TEC-2018

People

© Università degli Studi di Roma "La Sapienza" - Piazzale Aldo Moro 5, 00185 Roma