Modern human-engineered systems (e.g., power grid, communication, transportation, and smart building) with high degree of autonomy are becoming increasingly complex due to subsystem heterogeneity, uncertain operational & environmental dynamics, and decentralization of Decision & Control policies. Toward this direction, theories of control, communication and computation have matured in recent decades facilitating creation of cyber-physical systems, whose complexity is often hidden during nominal operations; however, it may become acutely conspicuous when contributing to rare cascading events, like the 2003 blackout in the power grid of North America. In these systems, the physical space serves as the source of information and the cyber space makes use of the generated information to make timely decisions for operation & control of the physical space. The presentation will address these issues and narrate the research experience in cyber-physical systems under several research grants.
The first part of the presentation will focus on the theory of Symbolic Dynamic Filtering (SDF) for abstraction of the events in the physical space as compressed information to be used in the cyber space. In this framework, sensor observations from a physical process entity are discretized both temporally and spatially to generate blocks of symbols, also called words that form a language. The grammar of a language is a set of rules that determine the relationships among the words to build sentences. A cyberphysical system is conjectured to be a linguistic source of information that is capable of generating a language for decision & control of the physical space. In this setting, a symbolic method of feature extraction and pattern classification has been developed upon the principles of signal preprocessing (e.g., wavelet transform) and probabilistic finite state automata (PFSA). The relational dependencies among heterogeneous sensors are modeled by cross PFSAs, from which low-dimensional feature vectors are generated for information fusion in real time over a sensor network. The proposed method has advantages of fast execution time and low memory requirements and is potentially well-suited for real-time implementation of the distributed decision & control laws within onboard sensor systems. In this context, several examples of validation will be presented based on laboratory experimentation data and field data.
The second part of the seminar will present an innovative concept of cognitive behavior prediction in cyber-physical systems. The objective is to coordinate human-machine collaboration such that human operators can assess and enable cyber-physical systems to utilize their experiential & unmodeled domain knowledge and perception for mission execution. The notion of quantum probability is proposed to construct a unified mathematical framework for interfacing between models of human cognition and machine intelligence. To this end, a biologically-inspired concept, based upon homeostasis and homeorhesis, is proposed for failure prognosis and self-healing in cyber-physical systems. Homeostasis in biological systems is the natural tendency of a living organism to maintain a stable macroscopic equilibrium, in the thermodynamic sense, among its interacting internal components, regardless of the interactions with the exogenous environment. Homeorhesis in biological systems is the natural tendency of a living organism to continue its development, under anticipated or unanticipated perturbations, albeit being possibly different from the original behavior. The goal is to conceive, formulate, and validate the concepts of homeostasis and homeorhesis to sustain order and normalcy in cyber-physical systems under both anticipated and unanticipated perturbations.
Asok Ray earned the Ph.D. degree in Mechanical Engineering from Northeastern University, Boston, MA, and also graduate degrees in each discipline of Electrical Engineering, Mathematics, and Computer Science. Dr. Ray joined the Pennsylvania State University at University Park, PA in July 1985, and is currently a Distinguished Professor of Mechanical Engineering and a Graduate Faculty of Electrical Engineering. Prior to joining Penn State, Dr. Ray held research and academic positions at Massachusetts Institute of Technology and Carnegie-Mellon University as well as research and management positions at GTE Strategic Systems Division, Charles Stark Draper Laboratory, and MITRE Corporation. Dr. Ray had been a Senior Research Fellow at NASA Glenn Research Center under a National Academy of Sciences award. Dr. Ray's research experience and interests include: Control and optimization of continuously varying and discrete-event dynamical systems; Intelligent instrumentation for real-time distributed systems; and Modeling and analysis of complex dynamical systems from thermodynamic perspectives in both deterministic and stochastic settings, as applied to Aeronautics and Astronautics, Undersea Vehicles and Surface Ships, Power and Processing plants, and Robotics.
Dr. Ray has authored or co-authored over five hundred research publications including over two hundred and sixty scholarly articles in refereed journals such as transactions of IEEE, ASME, and AIAA, and research monographs. Dr. Ray is a Fellow of IEEE, a Fellow of ASME, and a Fellow of World Innovative Foundation (WIF). Further details are available in the web address http://www.mne.psu.edu/Ray/ .