Cyber-physical applications often analyze collected sensor data using machine learning algorithms. Many existing sensing systems lack intelligence about the target and naively generate large-scale data, making communication and computation significantly costly. In many cases, however, the data generated by sensors only contain useful information for a small portion of the sensor activity. For example, machine learning algorithms continuously process the visual sensors used for environmental/security monitoring to detect sensitive activities. Still, these sensors only carry out useful information for a short time. On the other hand, biological sensors intelligently generate orders of magnitude less amount of data. This project develops machine learning algorithms that provide real-time feedback to sensors to ensure they only generate data needed for learning purposes. The approach is expected to provide up to four orders of magnitude data reduction from sensors. The results from this research will broadly impact many sensors used in internet-of-things applications, including infrastructure, mobile devices, autonomous systems, robotics, and healthcare. The project will also support underrepresented minority students through synergistic outreach plans and educational activities, including programs for K-12 students, undergraduate research opportunities, and new course development.

The research approaches introduced in this project aim to make fundamental changes to sensing systems in order to make future sensors intelligent for a wide range of cyber-physical applications. First, this project will develop novel brain-inspired learning algorithms that can provide fast and real-time feedback to the sensing module to intelligently control the rate of data generation from sensors. This feedback also makes sensors aware of the target task, enabling situational awareness. Second, the project will develop a novel framework that tightly integrates with a sensing circuit and brain-inspired algorithms to dynamically control the sensor functionality in a close-loop manner. The proposed hardware platform exploits the robustness of learning algorithms to design near-sensor computing platforms that are highly approximate, parallel, and efficient. Finally, this project aims to evaluate the effectiveness of the framework on multiple large-scale systems. The prototype will be fully released under an established open-source library for public dissemination.
 

This NSF CPS project aims to redesign the information structure utilized by system operators in today's electricity markets to accommodate technological advances in energy generation and consumption. The project will bring transformative change to power systems by incentivizing and facilitating the integration of non-conventional energy resources via a principled design of bidding, aggregation, and market mechanisms. Such integration will provide operators with the necessary flexibility to operate a network with high levels of renewable penetration. This will be achieved by a comprehensive bottom-down approach that will first identify the intrinsic cost of utilizing novel renewable resources and accommodate the operational ecosystem accordingly. The intellectual merits of the project include novel theories and algorithms for operating a vast number of distributed resources and testbed implementations of markets and controls. The project's broader impacts include K-12 and undergraduate programs, including in-class and extra-curricular STEM activities through, e.g., Hopkins in-class and extra-curricular STEM activities, and the Caltech WAVE summer research program.

Introducing distributed energy resources (DERs) at a large scale requires rethinking power grid operations to account for increased uncertainty and new operational constraints. The proposed research undertakes this task by overhauling the information structure that markets and grid controls utilize. We seek to characterize and shape how information is exchanged and used to manage the grid to improve efficiency, stability, and incentive alignment. The research is organized into three thrusts. Thrust 1 emphasizes the role of information in coordination. It seeks to characterize DER costs and constraints, designing bidding strategies tailored to convey information about the atypical characteristics of DER costs. Thrust 2 aims to develop aggregation strategies that efficiently manage resources by accounting for their cost and constraints, integrating DERs via an aggregate bid that protects sensitive user information and is robust to market manipulation. Finally, Thrust 3 characterizes the overall impact of DERs on operations. We will examine how user incentives that span across markets implicitly couple market outcomes and develop design mechanisms to mitigate inter-market price manipulation. We will also design pricing schemes that provide efficient DER allocation while preserving real-time operational constraints such as frequency regulation.

This award is funded in whole or in part under the American Rescue Plan Act of 2021 (Public Law 117-2).

This project aims to strengthen dependability and robustness of the electric power grid by improving the capability to aggregated power system state estimation (PSSE) methods to monitor and assess the health of a power grid. The electric power grid is a cyber-physical systems, essential for modern daily life that is always available on-demand nearly anywhere at any time. The grid is arguably the largest global engineered structure. The goals of this project are to (1) understand vulnerabilities intrinsic to traditional PSSE methods, and (2) improve the dependability and robustness of PSSE algorithms to potentially disruptive conditions. This project extends recently developed power system optimization techniques to enable better situation awareness of the operations of the overall power system. The project will work with the Arkansas State University?s outreach program ?P-20 Educational Innovation Center? to share this research and encourage careers in STEM fields.

This project extends convex relaxation-based techniques for large-scale cyber-physical power system in order to improve the monitoring, analysis, and controllability of these systems. The project presents an efficient convex relaxation-based approach to the PSSE problem. The project leverages semidefinite programming relaxation and extreme point approaches for large-scale alternating current (AC) power systems to provide tighter bounds on flow analyses than traditional PSSE. This project also presents an analysis showing the infeasibility of transitioning between certain operating regions in the power system that can be used to provide safety and security guarantees before initiating state transitions. This approach can identify specific types of sparse false data, that might arise from corrupted sensors by nature or design, and go undetected by current power systems' bad data detection algorithms. The proposed tighter relaxation schemes will be broadly applicable to a wide variety of nonconvex and nonlinear optimization problems in large flow-system models and in complex optimization problems.

The electric grid critical infrastructure is undergoing a major transformation from concentrated carbon-intensive legacy generation options to renewables in the form of distributed energy resources. The goal of this NSF workshop is to bring together a wide-spread collaboration among researchers from different scientific disciplines, such as data analytics, computational sciences, atmospheric sciences, and social and behavioral sciences ? addressing the engineering of complex systems will enable the convergent science needed to address these emerging challenges. The objective is to continue the discussion from the prior NSF-sponsored Workshops held in the USA in 2020 and 2021, and jointly with the European partners in Europe in 2022, attended by over 100 researchers from well over 50 US/European academic, government and industry organizations to further grid resilience discoveries and strategies through the proposed workshop with the participation of a wider scientific community.

The workshop will engage researchers from different scientific disciplines and a variety of application domains to address scientific, engineering, social, and economic challenges of interdependencies among critical infrastructures and how to inform the policy and regulatory process of the needed changes. The scientific merit is in the convergent scientific discussion among researchers and practitioners that will explore five scientific areas: (1) Data and physics-based modeling discovering new fundamentals in deep-learning approaches, (2) Transformational electric grid distributed control strategies laying the foundation for a resilient net-zero grid of the future, (3) Synergies between social and behavioral sciences to assess a human aspect of grid modernization leading to new models for electricity markets and incentives, and (4) Scalability of cybersecurity and privacy requirements across millions of internet-of-things consumers, and (5) Cross-dependency between electric grid infrastructure and other critical infrastructures. The workshop will also address how to develop partnerships that will work closely on informing and engaging the public and enhancing the STEM education and training of K-12 students and early-carrier professionals. The international experience of interacting with peers from 15 universities from a dozen leading European countries will be invaluable for the USA participants in forming a broader social, cultural, and political understanding of grid modernization.
 

Cancers are among the leading causes of death around the world, with an estimated annual mortality of close to 10 million. Despite significant efforts to develop effective cancer diagnosis and therapeutics, the ability to predict patient responses to anti-cancer therapeutic agents remains elusive. This is a critical milestone as getting the right choice of therapy early can mean superior anti-tumor outcomes and increased survival, while the wrong choice means tumor relapse, development of resistance, side effects without the desired benefit, and increased cost of treatment. An cyber-physical system that allows an accurate prediction of patient tumor responses to anti-cancer therapies; that is, enable real-time precision medicine, can have a transformative effect not only on health outcomes, but also on the costs of treatment. The goal of this project is therefore to develop an engineered cyber-physical system that combines advanced biological models with state-of-the-art artificial intelligence methods for predictive, automated screening of anti-cancer drugs and optimizations of their dosing. This will move science towards realizing the long-desired precision medicine paradigm leading to significant social impacts. The project has additional social impacts, including minimizing the exponentially growing ethical issues surrounding the use of animals in the past years through increased adoption of the engineered human cancer and heart tissue model systems. The project will provide opportunities to promote STEM education for K-12 students, train students, especially those from under-represented groups, and disseminate science and engineering knowledge to the public.

The investigators will leverage their expertise in biofabrication, tissue engineering, microfluidics, bioanalysis, and artificial intelligence to develop a generalized, self-dose-optimizing "multi-sensor-integrated multi-organ-on-a-chip" platform, which can be used to accurately predict both efficacy and safety of anti-cancer regimens in this project. The first innovation is the adoption of three-dimensional bioprinting for generating the vascularized ductal carcinoma model and vascularized cardiac tissue model, leading to the construction of a truly biomimetic human myocardium for evaluating drug toxicity. The adaptation of both of the bioprinted models to microfluidic systems is also a major innovation. Additionally, the real-time yet non-invasive monitoring of key biophysicochemical parameters will generate large-scale multi-dimensional data to enable accurate data-driven predictive modeling. Moreover, the platform will enable self-dose-optimization on the chips through a novel joint Bayes modeling implemented by two deep learning models capable of addressing multiple-instance learning, and dependency in sequences of multi-dimensional data, respectively. The project will use a range of commercially available cells to construct models and pursue the initial platform development and optimizations. Extensions are anticipated for human specimens in future iterations and other cancer treatment, drug combination, and dose optimization in anti-cancer regimens as a rapid and safe testing-bed.

Automating perennial farming operations in tree fruit crops is crucial for improving farming effectiveness, efficiency, and crop yield. However, current automation technologies lack full autonomy and are inefficient in complex farm environments. To address these challenges, our project aims to develop a cyber-physical system called Smart Harvesting. This system, integrating human intelligence and machine learning, will enhance decision-making and actuation, improving picking efficiency and system autonomy. By integrating Smart Harvesting into the crop production feedback loop, we will enrich the system's repertoire and reduce uncertainties in crop production. Additionally, the research outcomes can benefit other labor-intensive orchard operations like flower thinning and pruning, which also face labor shortage issues. This multidisciplinary research initiative will provide valuable opportunities for graduate and undergraduate students, particularly those from Hispanic and Native-serving institutions. The final product, a collaborative human-machine system for apple harvesting, will have a notable impact on rural agricultural communities. Its widespread adoption will contribute significantly to sustaining the competitiveness of the US tree fruit industry.

The project consists of three main areas of research. The first area focuses on creating a virtual reality orchard environment that is updated in real-time. This environment will use a network of sensors and a system called the Robotic Operation System that connects humans with machines. This will allow the control center to receive up-to-date 3D information about the orchard remotely. The second area aims to develop a collaborative framework where humans and machines work together effectively to harvest apples. This framework will utilize the virtual reality environment created in the first area. Human operators or machine learning techniques will be able to assist the robot system from a remote location. They can help the robot address challenges in apple picking, such as finding unidentifiable apples and determining the best way to retrieve them. The third area involves creating a constantly updating repertoire that incorporates information from human expertise and its own machine learning experience. It will record valuable information from human operators and its machine learning and use it to handle similar cases in the future autonomously. This repertoire will improve the performance of the apple harvesting robot, leading to better crop yield and quality.

Internet-of-Things (IoT) enabled manufacturing systems form a particularly important class of cyber-physical systems (CPS). IoT-enabled manufacturing systems have a physical fabric woven from a heterogeneous mix of machines carrying and processing materials across the factory floor. The cyber fabric for these systems is a heterogeneous mix of wired and wireless digital communication networks enabling the global visibility of the data streams used to manage the physical fabric?s workflows. These IoT-enabled systems are complex CPS with a great deal of modeling uncertainty. The physical and cyber fabrics are open to an external environment that can shift in an abrupt and unpredictable manner. Such shifts may be due to changes in customer work orders or due to environmental changes that cause traffic congestion in the cyber fabric?s wireless networks. The dynamics of both fabrics are coupled since congestion in the physical fabric may create congestion in the cyber fabric and vice versa. This complexity and uncertainty stand as major obstacles to the broader acceptance of IoT technologies by U.S. manufacturers. To lower the risk in adopting IoT technologies, this project proposes developing meta-learning methods that learn how to control complex CPS found in IoT-enabled manufacturing. This project will develop algorithms and software implementations of the meta-learning approach to controlling CPS. The project will benchmark the method?s performance on a testbed capturing the complex interactions between an IoT-manufacturing system?s physical and cyber fabrics.

This project uses meta-learning algorithms for the control of complex and uncertain cyber-physical systems. The approach adopts a new type of machine learning model called a behaviorally ordered abstraction (BOA). This model has greater cross-task generalization capacity, better sample efficiency, and greater interpretability than other deep learning methods. This modeling approach allows the project to address issues regarding the robust stability of deep reinforcement learning by embedding meta-learning in a generalized regulator that learns ?how? to configure controller synthesis across all tasks. This project will evaluate the project?s ?learning-how-to-control? framework on a multi-robotic testbed mimicking the use of WIFI connected robots moving materials across a factory floor. The project will investigate how to transfer the models and policies learned on the testbed to IoT-enabled factories found in local manufacturing facilities.

This NSF Smart and Connected Communities Integrative Research Grant (SCC-IRG) aims to address important equity and system integration challenges in mobility systems that could directly affect individual users' quality of life and access to critical services and employment opportunities. Results from this project will support the improvement of metropolitan areas broadly and the San Diego region specifically by exploiting emerging technologies and the public policy levers these technologies engender. The research team will work with transportation modelers and planners at the San Diego Association of Governments to develop a substantially improved decision support system for regional planning and investment decisions that could lead to more equitable, sustainable, and resilient future mobility system, such as solutions that could better connect people, especially disadvantaged populations, to jobs, healthcare, groceries, and other activities. The project also presents education and outreach opportunities to train next-generation engineers and practitioners in an integrated and multi-disciplinary research environment and broaden participation in STEM field.

To address socio-technical challenges related to equitable mobility, accessibility, and environmental sustainability, the research team will implement targeted improvements to regional transportation system models in the short term and fundamentally revamp regional transportation system models in the long run. To improve the models in the short run, the research team will develop flexible and detailed models of mobility-on-demand services, identify and develop equity metrics and equity analysis techniques, and develop low-resolution models for rapid analysis of potential policies. To fundamentally revamp regional models in the long run, the research team will develop a prescriptive (i.e., optimization-based) multi-level, multi-resolution, multi-objective pathway-based modeling framework to not only analyze but actually recommend combinations of transportation and land use policies and infrastructure investments over time. Moving from predictive to prescriptive modeling for regional transportation planning will represent a major theoretical contribution, as will incorporating equity into a multi-objective optimization problem formulation. Additionally, developing a multi-resolution modeling framework to support the bi-level, multi-objective prescriptive modeling framework will represent a valuable methodological contribution. Similarly, the new models provide sufficient flexibility to capture the important components of mobility-on-demand services and new technologies like connected automated vehicles thus represent an important methodological contribution that will speed the effective deployment of smart mobility solutions to address pressing social equity, sustainability, and economic challenges.

This Cyber-Physical Systems (CPS) project aims at designing theories and algorithms for scalable multi-agent planning and control to support safety-critical autonomous eVTOL aircraft in high-throughput, uncertain and dynamic environments. Urban Air Mobility (UAM) is an emerging air transportation mode in which electrical vertical take-off and landing (eVTOL) aircraft will safely and efficiently transport passengers and cargo within urban areas. Guidance from the White House, the National Academy of Engineering, and the US Congress has encouraged fundamental research in UAM to maintain the US global leadership in this field. The success of UAM will depend on the safe and robust multi-agent autonomy to scale up the operations to high-throughput urban air traffic. Learning-based techniques such as deep reinforcement learning and multi-agent reinforcement learning are developed to support planning and control for these eVTOL vehicles. However, there is a major challenge to provide theoretical safety and robustness guarantees for these learning-based neural network in-the-loop models in multi-agent autonomous UAM applications. In this project, the researchers will collaborate with committed government and industry partners on the use-case-inspired fundamental research, with a focus on promoting safety and reliability of AI, machine learning and autonomy in students with diverse backgrounds.

The technical objectives of this project include (1) Safety and Robustness of Single-Agent Reinforcement Learning: in order to address the ?safety critical? UAM challenge, the PIs plan the min-max optimization for single agent reinforcement learning to formally build sufficient safety margin, constrained reinforcement learning to formulate safety as physical constraints in state and action spaces, and the novel cautious reinforcement learning that uses variational policy gradient to plan the safest aircraft trajectory with minimum distributional risk; (2) Safety and Robustness of Multi-Agent Reinforcement Learning: in order to address the ?heterogeneous agents and scalability? challenge, a novel federated reinforcement learning framework where a central agent coordinates with decentralized safe agents to improve traffic throughput while guaranteeing safety, and a scaling mechanism to accommodate a varying number of decentralized aircraft; (3) Safety and Robustness from Simulations to the Real World: in order to address the ?high-dimensionality and environment uncertainty? challenge, the researchers will focus on the agents? policy robustness under distribution shift and fast adaptation from simulation to the real world. Specifically, value-targeted model learning to incorporate domain knowledge such as the aircraft and environment physics, and a safe adaptation mechanism after the RL model is deployed online for flight testing or execution is planned.

Origami-inspired structures that fold flat sheets along creases with designed patterns to create transformable structures have been widely applied in science and engineering, especially in space operations, e.g., for deployment of folded solar panels equipped on launched satellites. Although the deformation process plays an essential role in transitions between the origami states, few studies focus on the control and actuation of the origami folding mechanism toward high autonomy of the deformation process. This project aims to develop an autonomous origami-inspired transformable system to enable high-performance deformation maneuvering in space operations requiring frequent and/or time-responsive shape changes. The integrative research incorporating theory, analysis, algorithm development, and experimental verification will contribute to a theoretical and experimental platform to advance the autonomy of origami system operations in challenging environments. The research products will have significant impacts on the proliferated satellite marketplace where low mass, small volume, and adaptable structures/subsystems of space vehicles are in demand. Going beyond the applications in space missions, origami-inspired transformable systems have much broader applications in science and engineering. Moreover, the collaboration of experts in both cyber and physical areas promotes the creation of interdisciplinary products that bridge different disciplines.

To achieve the research goal of advancing autonomy of origami-inspired transformable systems, four research thrusts are identified, namely (1) developing a network-based approach for modeling and design of multi-shape origami structures, (2) designing an integrated sensing and control strategy with guaranteed controllability, reachability, and energy efficiency, (3) developing programmable untethered actuation via thermal loading to realize designed control maneuvers, and (4) evaluating the performance of autonomous systems using multiple origami structures in space operation missions. These identified research thrusts will together contribute to an analytical and computational framework for achieving autonomy of the origami deformation process, which will result in real-world applications in future space missions. Theoretically, the fundamental analysis based on networked control and graph modeling can lead to rigorous support of control performance in terms of controllability, reachability, and energy efficiency for the origami deformation process. Practically, the development of programmable untethered actuation enables the generation of designed control commands under operational constraints.