Our paper on ``Federated learning of predictive models from federated Electronic Health Records''...
CPS: Synergy: Collaborative Research: A Cyber-Physical Infrastructure for the “Smart City”
Funding Agency: National Science Foundation (NSF), Division for Computer and Network Systems, Cyber-Physical Systems (CPS).
Award Number: NSF CNS-1239021.
Principal Investigators: Christos Cassandras, Yannis Paschalidis, Azer Bestavros, and Assaf Kfoury at Boston University (with R. Gao at the Univ. of Conencticut and W. Gong at UMass Amherst).
Intellectual Merit. Our vision of a “smart city” encompasses well-managed and safer processes such as traffic control, efficient services such as parking, and new, innovative urban activities such as recharging electric vehicles. To enable these capabilities, we will study the components needed to establish a Cyber-Physical Infrastructure (CPI) for urban environments and address fundamental problems that involve data collection, dynamic resource allocation, real-time decision making, safety, and security, with emphasis on a balanced understanding of both “physical” and “cyber” components. Accordingly, our research is organized along two main directions capitalizing on the complementary PI skills and their synergistic potential: (i) Sensing and data acquisition using a new mobile sensor network paradigm designed for urban environments and consisting of both stationary and mobile nodes. This is the first step in enabling the CPI, so that data are reliably and securely communicated to control points and control actions are transmitted back to physical CPI components to “close the loop.” (ii) Decision Support for the “Smart City”, the next step in which formal verification and certification methods will be developed to capture and manipulate physical properties of the CPI, while innovative dynamic optimization techniques will be used for decision making and resource allocation purposes. This work will bring together and build upon the methodological advances the PIs have made spanning: optimization under uncertainty, computer simulation, discrete event and hybrid systems, cooperative control and games, stochastic supply chain logistics, system security, and formal verification and safety.
Target Applications. We will develop three specific “smart city” applications to demonstrate the feasibility of our vision and enhance our understanding of the CPI capabilities and limitations: (i) A “Smart Parking” system where parking spaces are optimally assigned and reserved for vehicles. Part of this system is already under development by the PIs at Boston U. (ii) A system for coordinating electric vehicles with battery charging stations that interface with a “smart grid.” (iii) Vehicular traffic regulation through dynamic traffic light control to achieve traditional as well as novel city-wide performance objectives. These applications are intended to demonstrate how to “close the loop” by using a CPI to not only provide information, but also make optimal decisions and implement them through proper actuation mechanisms.
Test Beds. An integral part of the proposed project is the application of our ideas and methods to a living laboratory in the Boston Back Bay neighborhood, an ongoing collaborative effort, termed Sustainable Neighborhood Laboratory (SNL), between Boston U., the Neighborhood Association of the Back Bay, commercial groups in the area, the City of Boston, and the local electricity distribution utility. This unique research and learning environment considers all aspects of the neighborhood as an “ecosystem” that offers insights into creating sustainable urban environments that support growth and improved quality of life while reducing environmental risks. At Boston U., a parking facility is already partially instrumented and will be fully equipped to implement our proposed “Smart Parking” system, which will also be deployed on-street in collaboration with the SNL and the City of Boston.
Broader Impact. Our “smart city” focus has the potential of revolutionizing the way we view the city in the future: from a passive living and working environment to a highly dynamic one with new ways to deal with transportation, energy, and safety. By teaming up with the stakeholders in the Boston SNL, we expect to also establish new collaborative models between universities and urban groups for cutting-edge research embedded in the deployment of an exciting unprecedented technological, economic, and sociological development. Our research scope clearly transcends the cyber-physical “smart city” realm and advances the state-of-the-art in all domains that rely on data collection using sensor networks, decision making in highly dynamic and uncertain environments, safety, and security. On the educational front, our plans include new courses, training graduate students, engaging undergraduate students, creating interactive educational software and demos, establishing summer internships, and reaching out to high school students through programs the PIs are involved with. Dissemination plans include capitalizing on the BU SNL and the “Sensor Network Consortium” established by the PIs, and organizing an academic workshop on the “smart city” and its ramifications.