Smart cities and Smart Grids are two exemplars of large scale cyber-physical systems that operate across geographically dispersed environments. With the increasing integration of edge devices, commodity computers, and low power sensors in these domains, we are seeing the emergence of cyber physical systems that operate with heterogeneous computing resources, which are distributed across physical environments. Our research group is exploring these societally important systems focusing on the following research challenges.
Development and Operations Management - Traditionally, cyber physical systems have been developed to operate with tightly controlled processes so that the design, analysis, and deployment can be done offline, before the system is deployed. As we think about modern operations that include over the air updates and flexible functionality that are implemented in software, the devops mechanisms used to design enterprise systems are now being used for CPS design. The challenge we face is – how can we ensure that the software engineering principles applied to enterprise structures can be used to guarantee safety and reliability for embedded software that have to operate under the physical constraints imposed by CPS components?
System Integration - Our exemplars represent large cyber physical system of systems that pose interesting integration and management challenges. For example, with the advancement of networking concepts we are quickly moving into a world where software applications from different CPS have to communicate with each other. This is a difficult challenge, especially if the systems are not properly orchestrated, the brittleness of their interfaces can lead to faults that can then propagate through the systems to cause failures.
System Reliability - System monitoring, prognostics, and fault detection and isolation are big challenges that we face as we scale up systems in size and complexity. Any system that operates over long periods of time has to cope with degradation associated with ageing, operational stress, and environmental conditions, that can result in failures of the associated physical components. Failures and latent bugs in the software add another source of degradation and failure, resulting in unoperational and compromised systems. It remains a challenge to understand and monitor degradation and failure caused by interactions between different subsystems of a large CPS that may operate across multiple physical domains.
Computation Platforms - A computation platform must be designed to accommodate and integrate heterogeneous components, operate at multiple time-scales (e.g., real time, near real time, and , long-term), allow for dynamic resource allocation, while accommodating a variety of topologies including edge networks, and providing, safety, reliability, and security guarantees. Our goal is to implement a stringent layered architecture that ensures that the layers interact across safe interface sets, provide efficiency guarantees, and ensure that the faults from a layer propagates across the layer in a way that allows better characterization of failure dynamics.
Big Data Analytics and Data Science - Systems such as the grid and Smart Cities are becoming more and more complex, and analysis of these systems using pure model-based approaches is becoming infeasible. On the other hand, the proliferation of data, storage and retrieval mechanisms, the availability of large and distributed computing resources, and the advances in analytics, machine learning, and mining methods is providing new opportunities to understand and act on the system operations. Our challenges here are to develop repositories and tool boxes that support a wide variety of analytics and mining routines, as well as work to address specific Smart City problems.
Abhishek Dubey is an assistant professor in the department of computer science and computer engineering at Vanderbilt University. He is also a senior research scientist at the Institute for Software Integrated Systems at Vanderbilt University. His research interests predominantly lies in resilient cyber-physical systems, including fault diagnositcs and prognostics and performance management algorithms. He is particularly interested in applying his work to solve the interoperability, scalability and fault management challenges faced by data intensive applications developed and deployed for smart and connect communities. His current projects are related to transporation, smart grid and emergency response domains. Abhishek completed his PhD in Electrical Engineering from Vanderbilt University in 2009. He received his M.S. in Electrical Engineering from Vanderbilt University in August 2005 and completed his undergraduate studies in electrical engineering from the Indian Institute of Technology, Banaras Hindu University, India in May 2001. He is a senior member of IEEE.
Prof. Biswas conducts research in Intelligent Systems with primary interests in hybrid modeling, simulation, and analysis of complex embedded systems, and their applications to diagnosis, prognosis, and fault-adaptive control. As part of this work, he has worked on fault diagnosis and fault-adaptive control of secondary sodium cooling systems for nuclear reactors, automobile engine coolant systems, fuel transfer systems for aircraft, Advanced Life Support systems and power distribution systems for NASA. He has also initiated new projects in health management of complex systems, which includes online algorithms for distributed monitoring, diagnosis, and prognosis. More recently, he has been applying data-driven methods for applications to Smart City projects. He is currently the lead for the trans-institutional program, VISOR – Vanderbilt Initiative for Smart-Cities Operations and Research, working closely with Metro Nashville agencies on a number of projects. He is a fellow of IEEE.
Chinmaya Samal is a graduate student in the department of computer science and computer engineering at Vanderbilt University. He also works as research assistant at the Institute for Software Integrated Systems at Vanderbilt University. He completed his undergraduate studies in Information Technology from Veer Surendra Sai University of Technology, India in May 2016.
Geoffrey Pettet is a graduate student in the Department of Computer Science and Computer Engineering at Vanderbilt University, and works as a research assistant at the Institute for Software Integrated Systems. He completed his undergraduate studies in computer science at Vanderbilt University in May 2016.
Fangzhou Sun is currently a Ph.D. student in computer science at Vanderbilt University. He received his M.S. degree in computer science from Vanderbilt University in 2015 and completed his undergraduate studies in computer science from Nanjing University, China in 2013. His main research topics include: (1) developing and managing applications, analytics tool boxes and platforms for smart city; (2) creating and integrating cyber-attack detection systems for heterogeneous web-based applications. He is also an active iOS app developer and web developer.
This project addresses the problem of urban transportation and congestion by directly engaging individual commuters. Because of the widespread use of smart devices, users are modeled as active agents in a shared economy, with algorithms designed to incentivize them to take actions that are efficient for the overall transportation system. Many commercially available Internet of Things solutions for multimodal transit focus on what is best for each individual from his or her local perspective. As the number of these local solutions grows, the misalignment between objectives of individual and the overall system also grows. An information bottleneck also forms, since massive data is collected by municipalities and users, but neither has the resources to develop real-time analytics and controls. Currently, very little has been done to provide an overarching solution that balances the needs of multiple parties, including commercial companies, municipal service providers, and individuals. The project will configure a computing and information sharing platform that overcomes the incentive gap between individuals and municipalities. This platform offers mixed-mode routing suggestions and general system information to travelers and in turn provides service providers with high-fidelity information about how users are consuming transportation resources.
In this project, we use the public transit system in the city of Nashville as a case study to develop tools and techniques for collecting the data, modeling and then analyzing these systems. The outcome of this project will be a smart phone application powered by a real-time decision support system that will enable the transit customers to engage more effectively with the system and allow the Metro transit authority to gain a better insight into several key aspects of the system, allowing them to make it more efficient. [Extended Abstract]
The CHARIOT (Cyber-pHysical Application aRchItecture with Objective-based reconfiguraTion) project, aims to address the challenges stemming from the need to resolve various challenges within extensible CPS found in smart Cities. CHARIOT is an application architecture that enables design, analysis, deployment, and maintenance of extensible CPS by using a novel design-time modeling tool and run-time computation infrastructure. In addition to physical properties, timing properties and resource requirements, CHARIOT also considers heterogeneity and resilience of these systems. The CHARIOT design environment follows a modular objective decomposition approach for developing and managing the system. Each objective is mapped to one or more data workflows implemented by different software components. This function to component association enables us to assess the impact of individual failures on the system objectives. The runtime architecture of CHARIOT provides a universal cyber-physical component model that allows distributed CPS applications to be constructed using software components and hardware devices without being tied down to any specific platform or middleware. It extends the principles of health management, software fault tolerance and goal based design. [Extended Abstract]
The future of the Smart Grid for electrical power depends on computer software that has to be robust, reliable, effective, and secure. This software will continuously grow and evolve, while operating and controlling a complex physical system that modern life and economy depends on. The project aims at engineering and constructing the foundation for such software: a 'platform' that provides core services for building effective and powerful apps, not unlike apps on smartphones. The platform will be designed by using and advancing state-of-the-art results from electrical, computer, and software engineering, will be documented as an open standard, and will be prototyped as an open source implementation.
Reliable operation of cyber-physical systems (CPS) of societal importance such as Smart Electric Grids is critical for the seamless functioning of a vibrant economy. Sustained power outages can lead to major disruptions over large areas costing millions of dollars. Efficient computational techniques and tools that curtail such systematic failures by performing fault diagnosis and prognostics are therefore necessary. The Smart Electric Grid is a CPS: it consists of networks of physical components (including generation, transmission, and distribution facilities) interfaced with cyber components (such as intelligent sensors, communication networks, and control software). In this this project we are developing new methods to build models for the smart grid representing the failure dependencies in both physical and cyber components. These models will be used to build an integrated system-wide solution for diagnosing faults and predicting future failure propagations that can account for existing protection mechanisms. The original contribution of this work is in the integrated modeling of failures on multiple levels in a large distributed cyber-physical system and the development of novel, hierarchical, robust, online algorithms for diagnostics and prognostics. [Extended Abstract].
1025 16th Ave S, Suite 102 Nashville, TN 37212
tel: +1 (615) 343-7472
fax: +1 (615) 343-7440