Processors in cyber-physical systems are increasingly being used in applications where they must operate in harsh ambient conditions and a computational workload which can lead to high chip temperatures. Examples include cars, robots, aircraft and spacecraft. High operating temperatures accelerate the aging of the chips, thus increasing transient and permanent failure rates. Current ways to deal with this mostly turn off the processor core or drastically slow it down when some part of it is seen to exceed a given temperature threshold. However, this pass/fail approach ignores the fact that (a) processors experience accelerated aging due to high temperatures, even if these are below the threshold, and (b) while deadlines are a constraint for real-time tasks to keep the controlled plant in the allowed state space, the actual controller response times that will increase if the voltage or frequency is lowered (to cool down the chip) are what determine the controlled plant performance. Existing approaches also fail to exploit the tradeoff between controller reliability (affected by its temperature history) and the performance of the plant. This project addresses these issues. Load-shaping algorithms are being devised to manage thermal stresses while ensuring appropriate levels of control quality. Such actions include task migration, changing execution speed, selecting an alternative algorithm or software implementation of control functions, and terminating prematurely optional portions of iterative tasks. Validation platforms for this project include automobiles and unmanned aerial vehicles. These platforms have been chosen based on both their importance to society and the significant technical challenges they pose. With CPS becoming ever more important in our lives and businesses, this project which will make CPS controllers more reliable and/or economical has broad potential social and economic impacts. Collaboration with General Motors promotes transition of the new technology to industry. The project includes activities to introduce students to thermal control in computing, in courses spanning high-school, undergraduate and graduate curricula.

The goal of the project is the development of the theory, hardware and computational infrastructure that will enable automatically transforming user-defined, high-level tasks such as inspection of hazardous environments and object retrieval, into provably-correct control for modular robots. Modular robots are composed of simple individual modules; while a single module has limited capabilities, connecting multiple modules in different configurations allows the system to perform complex actions such as climbing, manipulating objects, traveling in unstructured environments and self-reconfiguring (breaking into multiple independent robots and reassembling into larger structures). The project includes (i) defining and populating a large library of perception and actuation building blocks both manually through educational activities and automatically through novel algorithms, (ii) creating automated tools to assign values to probabilistic metrics associated with the performance of library components, (iii) developing a grammar and automated tools for control synthesis that sequence different components of the library to accomplish higher level tasks, if possible, or provide feedback to the user if the task cannot be accomplished and (iv) designing and building a novel modular robot platform capable of rapid and robust self-reconfiguration. This research will have several outcomes. First, it will lay the foundations for making modular robots easily controlled by anyone. This will enrich the robotic industry with new types of robots with unique capabilities. Second, the research will create novel algorithms that tightly combine perception, control and hardware capabilities. Finally, this project will create an open-source infrastructure that will allow the public to contribute basic controllers to the library thus promoting general research and social interest in robotics and engineering.

The objective of this proposal is to develop a distributed algorithmic framework, supported by a highly fault-tolerant software system, for executing critical transmission-level operations of the North American power grid using gigantic volumes of Synchrophasor data. As the number of Phasor Measurement Units (PMU) increases to more than thousands in the next 4-5 years, it is rather intuitive that the current state-of-the-art centralized communication and information processing architecture of Wide-Area Measurement System (WAMS) will no longer be sustainable under such data-explosion, and a completely distributed cyber-physical architecture will need to be developed. The North American Synchrophasor Initiative (NASPI) is currently addressing this architectural aspect by developing new communication and computing protocols through NASPI-net and Phasor Gateway. However, very little attention has been paid so far to perhaps the most critical consequence of this envisioned distributed architecture "namely", distributed algorithms, and their relevant middleware. Our primary task, therefore, will be to develop parallel computational methods for solving real-time wide-area monitoring and control problems with analytical investigation of their stability, convergence and robustness properties, followed by their implementation and testing against extraneous malicious attacks using our WAMS-RTDS testbed at NC State. In particular, we will address three critical research problems "namely" distributed wide-area oscillation monitoring, transient stability assessment, and voltage stability monitoring. The intellectual merit of this research will be in establishing an extremely timely application area of the PMU technology through its integration with distributed computing and optimal control. It will illustrate how ideas from advanced ideas from numerical methods and distributed optimization can be combined into power system monitoring and control applications, and how they can be implemented via fault-tolerant computing to maintain grid stability in face of catastrophic cyber and physical disturbances. The broader impact of this project will be in providing a much-needed application of CPS engineering to advance emerging research on PMU-integrated next-generation smart grids. Research results will be broadcast through journal publications, jointly organized graduate courses between NC State and University of Illinois Urbana Champagne, conference tutorials and workshops. Undergraduate research for minority engineering students will be promoted via the FREEDM Systems Center, summer internships via Information Trust Institute (UIUC) and RENCI, and middle/high-school student mentoring through the NCSU Science House program.

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). This grant provides funding to develop new methods to build models for the smart grid representing the failure dependencies in the physical and cyber components. The 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 will be 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. If successful, the model-based fault diagnostics and prognostics techniques will improve the effectiveness of isolating failures in large systems by identifying impending failure propagations and determining the time to critical failures that will increase system reliability and reduce the losses accrued due to failures. This work will bridge the gap between fault management approaches used in computer science and power engineering that are needed as the grid becomes smarter, more complex, and more data intensive. Outcomes of this project will include modeling and run-time software prototypes, research publications, and experimental results in collaborations with industry partners that will be made available to the scientific community.

Accurate and reliable knowledge of time is fundamental to cyber-physical systems for sensing, control, performance, and energy efficient integration of computing and communications. This statement underlies the proposal. Emerging CPS applications depend on precise knowledge of time to infer location and control communication. There is a diversity of semantics used to describe time, and quality of time varies as we move up and down the system stack. System designs tend to overcompensate for these uncertainties and the result is systems that may be over designed, inefficient, and fragile. The intellectual merit derives from the new and fundamental concept of time and the holistic measure of quality of time (QoT) that captures metrics including resolution, accuracy, and stability. The proposal builds a system stack ("ROSELINE") that enables new ways for clock hardware, operating system, network services, and applications to learn, maintain and exchange information about time, influence component behavior, and robustly adapt to dynamic QoT requirements, as well as to benign and adversarial changes in operating conditions. Application areas that will benefit from Quality of Time will include: smart grad, networked and coordinated control of aerospace systems, underwater sensing, and industrial automation. The broader impact of the proposal is due to the foundational nature of the work which builds a robust and tunable quality of time that can be applied across a broad spectrum of applications that pervade modern life. The proposal will also provide valuable opportunities to integrate research and education in graduate, undergraduate, and K-12 classrooms. There will be extensive outreach through publications, open sourcing of software, and participation in activities such as the Los Angeles Computing Circle for pre-college students.

This project investigates a new type of cyber-physical system (CPS), comprising magnetic nanoparticles in a fluidic environment such as human tissue whose motion is controlled by a computer via a magnetic field. The research aims to develop computational and experimental tools to perform the dynamic modeling, closed loop control and experimental validation of such a system of nanoparticles under guidance and observation using a magnetic resonance imaging (MRI) environment. The envisioned CPS infrastructure is composed of a new computational platform to perform 3D simulation, visualization and post-processing analysis of the aggregation and disaggregation process of magnetic nanoparticles within a fluidic environment like the small arteries and arterioles or fluid-filled cavities of the human body. It also includes the development of robust control algorithms for the guidance of a swarm of magnetic nanoparticles in a MRI environment. Experimental validation is to be performed in clinical MRI scanners and in customized laboratory test-beds that generate controllable magnetic fields able to move magnetic nanoparticles in fluidic environments. Potential applications of this basic research include nano-robotic drug delivery systems, composed of a system of magnetic nanoparticles guided by MRI scanners for targeted drug delivery in the human body. The project integrates education through participation of graduate and undergraduate students in the research, and involvement of the PI and graduate students in several outreach activities for students in high and middle schools.

Accurate and reliable knowledge of time is fundamental to cyber-physical systems for sensing, control, performance, and energy efficient integration of computing and communications. This statement underlies the proposal. Emerging CPS applications depend on precise knowledge of time to infer location and control communication. There is a diversity of semantics used to describe time, and quality of time varies as we move up and down the system stack. System designs tend to overcompensate for these uncertainties and the result is systems that may be over designed, inefficient, and fragile. The intellectual merit derives from the new and fundamental concept of time and the holistic measure of quality of time (QoT) that captures metrics including resolution, accuracy, and stability. The proposal builds a system stack ("ROSELINE") that enables new ways for clock hardware, operating system, network services, and applications to learn, maintain and exchange information about time, influence component behavior, and robustly adapt to dynamic QoT requirements, as well as to benign and adversarial changes in operating conditions. Application areas that will benefit from Quality of Time will include: smart grad, networked and coordinated control of aerospace systems, underwater sensing, and industrial automation. The broader impact of the proposal is due to the foundational nature of the work which builds a robust and tunable quality of time that can be applied across a broad spectrum of applications that pervade modern life. The proposal will also provide valuable opportunities to integrate research and education in graduate, undergraduate, and K-12 classrooms. There will be extensive outreach through publications, open sourcing of software, and participation in activities such as the Los Angeles Computing Circle for pre-college students.

The goal of the project is the development of the theory, hardware and computational infrastructure that will enable automatically transforming user-defined, high-level tasks such as inspection of hazardous environments and object retrieval, into provably-correct control for modular robots. Modular robots are composed of simple individual modules; while a single module has limited capabilities, connecting multiple modules in different configurations allows the system to perform complex actions such as climbing, manipulating objects, traveling in unstructured environments and self-reconfiguring (breaking into multiple independent robots and reassembling into larger structures). The project includes (i) defining and populating a large library of perception and actuation building blocks both manually through educational activities and automatically through novel algorithms, (ii) creating automated tools to assign values to probabilistic metrics associated with the performance of library components, (iii) developing a grammar and automated tools for control synthesis that sequence different components of the library to accomplish higher level tasks, if possible, or provide feedback to the user if the task cannot be accomplished and (iv) designing and building a novel modular robot platform capable of rapid and robust self-reconfiguration. This research will have several outcomes. First, it will lay the foundations for making modular robots easily controlled by anyone. This will enrich the robotic industry with new types of robots with unique capabilities. Second, the research will create novel algorithms that tightly combine perception, control and hardware capabilities. Finally, this project will create an open-source infrastructure that will allow the public to contribute basic controllers to the library thus promoting general research and social interest in robotics and engineering.

This project is a modular, computationally-distributed multi-robot cyberphysical system (CPS) for assisting young developmentally-delayed children in learning to walk. The multi-robot CPS is designed to function in the same way as an adult assisting a child in learning to walk (addressing the research target area of a science of CPS by introducing developmental rehabilitation robotics). It addresses the research target area of new CPS technology by introducing a multi-robot system: 1) a multi-cable scaffold robot that continuously modulates the stabilization of medio-lateral and anterior-posterior sway, and 2) a soft, wearable, exosuit robot with embedded sensing and actuation, which assists with stance push off and swing flexion. The objective is to build a prototype multi-robot CPS and perform tests with human subjects to evaluate the CPS functionality, safety, and interoperability (addressing the research target area of engineering CPS). Longitudinal tests of typically developing and developmentally delayed children learning to walk with or without assistance of the multi-robot CPS are conducted in a motion capture laboratory. Body center of mass behavior as well as gait parameters of walking are measured as the two robots work together to assist the child in maintaining balance and propelling the body forward with each step. This exosuit/scaffolding multi-robot technology will advance knowledge within engineering with bio-inspired soft components, including miniature pneumatic artificial muscle actuators with embedded sensors that enable the control of the muscles in real time. The bio-inspired architecture and material components of the exosuit will make possible a new generation of ?smart fabric? that acts in concert with the body for efficient energy use. The exosuit is part of a larger modular design that makes it possible to couple it to additional assistive robots via a modular communications network. Together, the exosuit, scaffold robot, and wireless communications network for modular CPS, will advance knowledge for the engineering of other CPS that require high levels of interoperability and safety, such as medical CPS. The multi-robot CPS is designed for children who are developmentally delayed as a result of early brain injury. The long term consequences of early brain injury, e.g., in children born prematurely, constitute a major health problem and a significant emotional and financial burden for families and society. The use of a multi-robot cyberphysical system as part of a rehabilitation program may be able to harness the potential of the nervous system for plasticity, the ability to re-organize its structure, function, and connections. The focus is on young children with a history of early brain injury due to prematurity. However, this new cyberphysical system will have a much broader impact in restoring function throughout the life span. Neuroplasticity is not just an immediate response to injury, but occurs throughout the developmental period, providing an opportunity to promote repair and re-education, and restore function. A key to this broad application is the developmentally-motivated, modular structure and interoperability of the exosuit.

The objective of this research is to develop a comprehensive theoretical and experimental cyber-physical framework to enable intelligent human-environment interaction capabilities by a synergistic combination of computer vision and robotics. Specifically, the approach is applied to examine individualized remote rehabilitation with an intelligent, articulated, and adjustable lower limb orthotic brace to manage Knee Osteoarthritis, where a visual-sensing/dynamical-systems perspective is adopted to: (1) track and record patient/device interactions with internet-enabled commercial-off-the-shelf computer-vision-devices; (2) abstract the interactions into parametric and composable low-dimensional manifold representations; (3) link to quantitative biomechanical assessment of the individual patients; (4) facilitate development of individualized user models and exercise regimen; and (5) aid the progressive parametric refinement of exercises and adjustment of bracing devices. This research and its results will enable us to understand underlying human neuro-musculo-skeletal and locomotion principles by merging notions of quantitative data acquisition, and lower-order modeling coupled with individualized feedback. Beyond efficient representation, the quantitative visual models offer the potential to capture fundamental underlying physical, physiological, and behavioral mechanisms grounded on biomechanical assessments, and thereby afford insights into the generative hypotheses of human actions. Knee osteoarthritis is an important public health issue, because of high costs associated with treatments. The ability to leverage a quantitative paradigm, both in terms of diagnosis and prescription, to improve mobility and reduce pain in patients would be a significant benefit. Moreover, the home-based rehabilitation setting offers not only immense flexibility, but also access to a significantly greater portion of the patient population. The project is also integrated with extensive educational and outreach activities to serve a variety of communities.