Talks

Abstract: In modern technologies such as autonomous vehicles and service robots, control engineering plays a crucial role for the overall performance and safety of the system. However, the control design becomes often very time-consuming or infeasible due to the increasing complexity of recent technological advancements. The classical control approaches, which are based on models of the systems using first-order principles, are not satisfactory in the presence of complex dynamics, e.g., for highly nonlinear systems or interaction with prior unknown environment. Recent findings in computational intelligence and machine learning have shown that data-driven approaches lead to very promising results in a wide application range as they require only a minimal prior knowledge for the modeling of complex dynamics. Within the past two decades, Gaussian process (GP) models have been used increasingly as a data-driven technique due to many beneficial properties such as the bias-variance trade-off and the strong connection to Bayesian mathematics. However, the major drawback in data-driven approaches frequently manifests as unpredictable outcomes. Therefore, the current application of GP models in control scenarios is typically limited to non-critical and low performance systems due to their unpredictable "blackbox" behavior. In this talk, I will present our results on identification and control based on Gaussian process models. First, I analyze the control related properties of GP dynamical models which heavily depend on the underlying kernel function. As GP dynamical models generally lead to non-Markovian systems, I introduce approximations that achieve Markovian dynamics and show how to incorporate control theoretic prior knowledge into GP models. Afterwards, a GP model-based control law is presented which guarantees the safe control of electromechanical systems with unknown dynamics. I demonstrate how to actively exploit the uncertainty of the GP model to guarantee performance and stability of the closed-loop.

Bio: Thomas Beckers is a postdoctoral researcher at the Department of Electrical and Systems Engineering, University of Pennsylvania. In 2020, he earned his doctorate in electrical engineering at the Technical University of Munich (TUM), Germany. He received the B.Sc. and M.Sc. degree in electrical engineering in 2010 and 2013, respectively, from the Technical University of Braunschweig, Germany. In 2018, he was a visiting researcher at the University of California, Berkeley. His research interests include learning-based identification and control, nonparametric systems, and formal methods for safe learning.

Abstract: In many real-world applications, the system models employed in control are uncertain and inaccurate, e.g, due to unmodeled dynamics such as friction, complex environmental effects found in aerial and underwater robotics, or the lack of first principle models in human-robot interaction. In order to improve control performance despite the model inaccuracy, supervised machine learning is commonly used to update these models using measurements of the system. This approach raises questions regarding the stability of the closed-loop system and the impact of data on control error guarantees. In this talk, we investigate how Bayesian machine learning intuitively allows to address these problems through an explicit uncertainty representation, which allows the derivation of probabilistic prediction error bounds. These error bounds admit a pessimistic stability analysis of control laws employing the learned model, and can be used to determine the relationship between training data and control error bounds, such that a high data-efficiency in learning can be achieved. Finally, we illustrate how the derived theoretical guarantees can be realized with computationally efficient algorithms in practice.

Bio: Armin Lederer received the B.Sc. and M.Sc. degree in electrical engineering and information technology from the Technical University of Munich, Germany, in 2015 and 2018, respectively. Since June 2018, he has been a PhD student at the Chair of Information-oriented Control, Department of Electrical and Computer Engineering at the Technical University of Munich, Germany. His current research interests include the stability of data-driven control systems and probabilistic machine learning in closed-loop systems.

As the scale of networked control systems (NCS) increases and interactions between different subsystems become more sophisticated, questions of their resilience to failures increase in importance. Human intervention in those systems increases this complexity, especially when it comes in the form of an adversary. In this talk, we introduce game-theoretic approaches to an optimal sensor and actuator placement in networked control systems to counteract strategic adversarial actions. In particular, 1- in an attacker-defender game, a strategic selection of nodes (to add feedback loops) is discussed to mitigate the impact of the attack, 2- in an attacker-detector game, a strategic selection of nodes (to place sensors on) is discussed to detect the largest possible number of attacks. The attack impact in problem 1 is modeled by an appropriate system norm from the attack input to the states of the agents and the detection in problem 2 is modeled by the structural invertibility of the system. Using tools from systems and control, graph theory, and structured systems theory, we find game equilibria and provide graph-theoretic interpretations for those equilibria.

Bio: Mohammad Pirani is a senior research scientist with the Department of Electrical and Computer Engineering, University of Waterloo. Before that, he held postdoctoral research positions at the University of Toronto (from 2019 to 2021) and KTH Royal Institute of Technology, Sweden (2018 to 2019). He received his M.A.Sc. in Electrical and Computer Engineering and Ph.D. in Mechanical and Mechatronics Engineering, both from the University of Waterloo, Canada, in 2014 and 2017, respectively. His research interests include resilient and secure networked control systems with applications to intelligent transportation systems and multi-agent systems. He is a member of the IEEE-CSS technical committee on smart cities.

Abstract: Hybrid dynamical models are widely used to characterize processes that involve both discrete and continuous behaviors. Examples include contacts in robot locomotion and manipulation, traffic lights in urban vehicular networks, and protein-based switches in gene circuits. The combinatorial nature of such systems poses computational challenges for controller design both in real-time and offline settings. Furthermore, guaranteeing robustness of hybrid systems is difficult but of essence in many applications that model discrepancies are present. In this talk I will introduce polytopic trees, which is a method to compute controllers that consist of polytopic sets and control laws connected to each other forming a reachability tree. I describe a method to compute polytopic trajectories such that the nominal trajectory design, tracking controller, and its viable sets are all jointly optimized. The theoretical guarantees and the practical advantages of such joint optimization will be discussed. I will also provide the newest results on the mathematical foundations of polytope containment problems and its relevance to control problems. Throughout the talk, I will present illustrative examples including case studies in contact-aware robotics.

Bio: Sadra Sadraddini got his PhD in Mechanical Engineering from Boston University in 2018 and bachelors in Mechanical Engineering and Aerospace Engineering from Sharif University of Technology in 2013. His PhD thesis was titled "formal methods for resilient control" and provided optimization-based methods to formal design of controllers for hybrid systems such as transportation networks. He was a postdoctoral research associate in Robot Locomotion Group at Massachusetts Institute of Technology from 2018 to 2021. He is currently a senior robotics engineer at Dexai Robotics. His research brings formal methods and mathematical optimization to controller design and verification with an emphasis on applications in robotics.

Abstract: While obtaining an accurate model is often a time-consuming task requiring expert knowledge, data in form of input-output trajectories are often readily available. Therefore, there has been an increasing interest in direct systems analysis and control from data without first identifying a full mathematical model of the system. In this talk, two different approaches to data-driven representations of linear time-invariant (LTI) systems will be discussed and their relevance for data-driven systems analysis and controller design. For both viewpoints, we will present necessary and sufficient conditions for an unknown LTI system to satisfy certain dissipativity properties or integral quadratic constraints solely on the basis of data, which boil down to simple linear matrix inequalities (LMIs). We show that, even in the case of noisy data, rigorous guarantees for the respective system properties can be obtained. Numerical as well as experimental examples showcase the potential of the proposed methods. Finally, we link the results to other data-driven analysis and control results and discuss remaining open problems and future potential of the respective viewpoints.

Bio: Anne Koch (née Romer) received the B.Sc. degree in Engineering Cybernetics from the University of Stuttgart and holds both an M.Sc. in Engineering Science and Mechanics from the Georgia Institute of Technology and an M.Sc. in Engineering Cybernetics from the University of Stuttgart. She is currently a research and teaching assistant at the Institute for Systems Theory and Automatic Control pursuing a Ph.D. degree within the International Max Planck Research School for Intelligent Systems. She received the IFAC Young Author Award at the joint ROCOND and LPVS in 2018 and the Publication Prize of the University of Stuttgart in 2021. Her research interests include data-driven systems analysis and control.

Abstract: Situational awareness of the transient dynamics of power systems is becoming increasingly relevant as these systems undergo major changes due to the massive introduction of power-electronics-interfaced equipment, growing transit power flows and fluctuating (renewable) generation. These developments render the steady-state assumptions used in traditional static state estimation questionable. Thus, the deployment of methods for dynamic state estimation (DSE) is gaining increasing importance for power system control and protection. A key enabler for DSE is the growing availability of phasor measurement units (PMUs). By exploiting the huge potential provided by PMUs, in the present talk a decentralized mixed algebraic and dynamic state observation approach is introduced, which is suitable for multi-machine power systems with unknown inputs and partially known parameters. The effectiveness of the proposed DSE technique is demonstrated in extensive simulations based on the New England IEEE-39 bus system.

Bio: Johannes Schiffer studied Engineering Cybernetics at the University of Stuttgart, Germany, and Lund University, Sweden. Then he held appointments as research associate at the Chair of Sustainable Electric Networks and Sources of Energy (2009-2011) and the Control Systems Group (2011-2015) both at TU Berlin, Germany. In 2015, he received a Ph.D. degree (Dr.-Ing.) in Electrical Engineering from TU Berlin for a thesis on Stability and Power Sharing in Microgrids. From 2015-2018, he was a Lecturer (Assistant Professor) in Smart Energy Systems at the School of Electronic and Electrical Engineering, University of Leeds, UK. Johannes joined Brandenburg University of Technology Cottbus-Senftenberg (BTU) as Chair of Control Systems and Network Control Technology in 2018. In 2020, he also served as Deputy of Research at BTU. Together with his coworkers, he is a recipient of the Automatica Paper Prize over the years 2014-2016. His main research interests are in distributed control methods for complex and networked systems and their application to low-emission energy systems.

Abstract: Cyber-physical systems (CPS) are complex systems with intricate interaction between physical and computational components. These systems are often required to perform complex-logic tasks that are usually specified as linear temporal logic (LTL). Even though LTL can effectively describe many linear-time properties like safety, liveness, etc., that consider only single execution traces at a time, many information-flow and planning specifications such as opacity and optimality require the quantification between multiple execution traces and cannot be expressed by LTL. Such properties are called hyperproperties and can be specified by hyper-temporal logics (HyperLTL). This talk first introduces the syntax and semantics of hyperproperties specified as HyperLTL and provides an overview on some preliminary concepts relevant to the topic. It then discusses an automata-theoretic approach for the sound verification of continuous-state control systems against HyperLTL specifications. Specifically, this talk describes a multi-state procedure for the verification, involving (i) decomposition of HyperLTL specifications into simple conditional invariances by utilizing their corresponding automata, (ii) synthesis of so-called augmented barrier certificates (ABCs) to provide sufficient conditions for the satisfaction of conditional invariances, and (iii) utilization of conditional invariance guarantees to reason about the satisfaction of HyperLTL specifications.

Bio: Mahathi Anand is a PhD candidate in the Software and Computational Systems lab at LMU Munich since July 2019. She received her B.Tech (Electrical and Electronics Engineering) from SRM Institute of Science and Technology, India in 2016 and M.Tech (System and Control) from Indian Institute of Technology Roorkee, India in 2019. She also received an IIT-MSP scholarship from DAAD to undertake her master’s thesis as an exchange student at Technical University of Munich in 2018. Her research interests lie in the formal analysis of control systems against (hyper)properties using discretization-free approaches based on barrier certificates.

Abstract: The singular values of the linearization of a nonlinear system can be used to quantify various characteristics of the system, including its (exponential) stability behavior and the Lyapunov dimensions of its invariant sets. In the context of remote state estimation, the smallest capacity of a communication channel between the sensor and the estimator, above which the system can be observed robustly, can also be characterized in terms of these singular values. In general, the singular values of a linear operator depend on the inner products on domain and codomain of the operator. For nonlinear systems, a Riemannian metric on the state space assigns an inner product to each state, with respect to which the singular values of the linearized system can be computed. If the Riemannian metric is chosen carefully, and is well-adapted to the system, the asymptotic behavior of the system already becomes apparent from the singular values of the time-one-map or the generating vector field (in analogy to Lyapunov functions). In this talk, I present a subgradient algorithm on a space of conformal Riemannian metrics on the state set which minimizes partial sums of singular values. This allows to compute metrics which are useful for the design of observers in the remote state estimation problem. Slight modifications of the algorithm can be used to detect exponential stability and to estimate the Lyapunov dimension of invariant sets.

Bio: TBA

Abstract: Model predictive control (MPC) has become one of the most successful modern control concepts, mainly thanks to its ability to directly incorporate hard state and input constraints as well as some performance criterion into the controller design. In the last decades, various MPC schemes for linear and nonlinear systems have been proposed which allow for closed-loop stability, robustness, and performance guarantees. To this end, a reasonably well identified model of the considered system is needed. However, in some applications, obtaining such a model by (classical) system identification can be difficult or the physical modelling process might be expensive. In such cases, MPC schemes are of high interest which suitably employ collected data for predictions. In this talk, we discuss a first data-based MPC scheme for which rigorous closed-loop stability and robustness guarantees can be given. To this end, a trajectory-based system representation is used for prediction which allows to express all input/output trajectories of a (linear) system in terms of a single, sufficiently exciting, input/output trajectory. Based on this, we design data-based MPC schemes and derive closed-loop stability and robustness guarantees, where for the latter various connections between design parameters, data properties, and the resulting closed-loop behavior are revealed.

Bio: Matthias A. Müller received a Diploma degree in Engineering Cybernetics from the University of Stuttgart, Germany, and an M.S. in Electrical and Computer Engineering from the University of Illinois at Urbana-Champaign, US, both in 2009. In 2014, he obtained a Ph.D. in Mechanical Engineering, also from the University of Stuttgart, Germany, for which he received the 2015 European Ph.D. award on control for complex and heterogeneous systems. Since 2019, he is director of the Institute of Automatic Control and full professor at the Leibniz University Hannover, Germany. He obtained an ERC Starting Grant in 2020 and is recipient of the inaugural Brockett-Willems Outstanding Paper Award for the best paper published in Systems & Control Letters in the period 2014-2018. His research interests include nonlinear control and estimation, model predictive control, and data-/learning-based control, with application in different fields including biomedical engineering.

Abstract: An invariant set of a dynamical system refers to a region where the trajectory will never leave once it enters. It is widely used in systems and control for stability analysis and control design. In particular, many control methods for safety-critical applications exploit set-invariance arguments. In this talk, I will talk about generalization of the set invariance property to a model-free setting in which the system is a black box without a dynamical model. First, I will show a data-driven method for computing the maximal invariant set relying on the observation of trajectories. This method mimics a standard fixed-point algorithm in the white-box setting. Then, I will present probabilistic guarantees on set invariance by introducing almost invariant sets, which are invariant everywhere except a small subset. Finally, I will also show that the concept of almost invariant sets can be extended to hybrid systems, e.g., switched linear systems.

Bio: Zheming Wang is a postdoctoral research in the department of mathematical engineering at UCLouvain in Belgium. He is broadly interested in the area of control theory and optimization. His recent research interests focus on rigorous data-driven methods for analyzing and controlling complex systems with formal guarantees. He received a B.S. degree in Mechanical Engineering from Shanghai Jiao Tong University, China, in 2012, and a Ph.D degree in Mechanical Engineering from National University of Singapore in 2016.

Abstract: In this talk, I will discuss several topics on ADAS/AV development. The first part focuses on ADAS algorithm developments toward safety and comfort driving objectives. Some use cases developments on safety-critical collision avoidance, autonomous valet parking, and human-like autonomous driving will be shown. I will also discuss approaches how to combine model-based and data learning to enhance both safety and comfort. The second part concerns ADAS testing and validation. Examples on scenario generation, virtual and mixed virtual-real testing, and proving ground track testing will be demonstrated. The works are part of Siemens Digital Industries Software ADAS/AV solutions, in particular Simcenter Engineering Services.

Bio: Dr. Son Tong a senior research engineer and project manager at Siemens Digital Industries Software in Leuven, Belgium. His current main focuses are ADAS and autonomous vehicle topics, for example, motion planning, control, sensor fusion, virtual and physical testing, and safety and comfort analyses. He received his PhD as a Marie Curie fellow at KULeuven in the field of learning control in 2016. Son Tong is actively involved in different research and development activities in EU/Belgian projects, publishing, collaborations, and supervising industrial PhDs/researchers. He also serves in Conference Editorial Board of IEEE Control System Society. He was awarded 2018 Siemens PL Invention of the Year Award and in the finalist of the AutoSens Award 2019 in Most Influential Research.

Abstract: The increasing size and heterogeneity of modern engineering systems calls for a theory for their design and control that is inherently modular, i.e., is based on considering subsystems independently. In this talk, we will present such theory by introducing assume-guarantee contracts for linear dynamical systems that. These contracts can be regarded as specifications on dynamical systems. Namely, assumptions capture the knowledge about the environment in which the system will operate, whereas the guarantees specify the required dynamical behavior of the system when interconnected to its environment. Additionally, we develop notions of comparing contracts as well as composition of contracts, leading to a contract theory that enables modular design.

Bio: Bart Besselink is an assistant professor at the Bernoulli Institute for Mathematics, Computer Science and Artificial Intelligence of the University of Groningen, the Netherlands. He received the M.Sc. degree (cum laude) in Mechanical Engineering in 2008 and the Ph.D. degree in 2012, both from Eindhoven University of Technology, the Netherlands. He was a short-term visiting researcher at the Tokyo Institute of Technology, Japan, in 2012, and a post-doctoral researcher at the Department of Automatic Control and ACCESS Linnaeus Centre at KTH Royal Institute of Technology, Sweden, between 2012 and 2016. His main research interests include systems theory and model reduction for nonlinear dynamical systems and large-scale interconnected systems, with particular interested in modular approaches for system analysis and design. In addition, he works on applications in the field of intelligent transportation systems.

Abstract: Cyber-Physical Systems (CPS) are technical systems where a large software stack orchestrates the interaction of physical and digital components. Such systems are omnipresent in our daily life and their correct behavior is crucial. However, developing safe and reliable CPS is challenging. A promising direction towards this goal is the use of formal methods: automated methodologies that ensure system requirements during design-time. The main challenge in their application to CPS is the large amount of interacting heterogeneous components for which synthesis tools must automatically and locally generate code implementing a desired joint behavior.

In the first part of my talk, I will overview various techniques that we have recently developed to expand the scope of automated formal techniques for CPS design.

In the second part of my talk, I will focus on a particular technique, called abstraction-based controller design (ABCD), which allows to synthesize discrete controllers for continuous dynamical systems to enforce discrete temporal logic specifications. I will show limitations of existing ABCD techniques in the presence of output feedback. I will then present our recently developed algorithm that allows for sound and complete ABCD in this setting and compare it to existing techniques.

Bio: Anne-Kathrin Schmuck is an independent research group leader at the Max Planck Institute for Software Systems (MPI-SWS) in Kaiserslautern, Germany, funded by the Emmy Noether Programme of the German Science Foundation (DFG). She received the Dipl.-Ing. (M.Sc) degree in engineering cybernetics from OvGU Magdeburg, Germany, in 2009 and the Dr.-Ing. (Ph.D.) degree in electrical engineering from TUBerlin, Germany, in 2015. Between 2015 and 2020 she was a Postdoctoral researcher at MPI-SWS. She currently serves as the co-chair of the IEEE CCS Technical Committee on Discrete Event Systems and as associate editor for the Springer Journal on Discrete Event Dynamical Systems. Anne's current research interests include abstraction based controller design, reactive synthesis, supervisory control theory, hierarchical control and contract-based distributed synthesis.

Abstract: Cyber-physical systems (CPS) are complex systems resulting from the interaction between digital computational devices and the physical plants. Within CPS, (embedded) control software plays a significant role by monitoring and adjusting several physical variables, e.g. temperature, velocity, and pressure, through feedback loops where physical processes affect computation and vice versa. Increasing levels of autonomy in modern safety-critical CPS such as vehicles, airplanes, power plants, and medical robots, poses serious questions about their safety. Any failure in the safety-critical control software (SCCS) costs hundreds of lives. A recent example is the erroneous activation of the maneuvering characteristics augmentation system in Boeing 737 MAX airplanes causing the death of 346 passengers in two consecutive crashes in 2018 and 2019, respectively. Ensuring the correctness of SCCS is then very crucial.

In this talk, Khaled will introduce an end-to-end approach for designing foolproof SCCS by using unambiguous formal descriptions for design requirements and, at the same time, automating the design, development and implementation phases of SCCS lifecycle. He will first revise "Symbolic Control", an approach for automated synthesis of controllers for CPS that has become popular in the last few years. Then, he will discuss and address problems of symbolic control that hinder applying it to real world applications. More specifically, he will discuss two problems: (1) the complexity of symbolic control algorithms which increases exponentially with the size of systems, and (2) the absence of formal implementations of the synthesized controllers. Khaled will introduce parallel scalable algorithms of symbolic control and show that they lead to reductions in the computational complexity which allows for online real-time implementations. He will then discuss formal representations of the designed controllers and will introduce an automated approach for the implementation and code-generation of SCCS.

Bio: Mahmoud Khaled is a PhD candidate in the Department of Electrical and Computer Engineering at the Technical University of Munich (TUM), Germany, and a research assistant in the Chair of Software and Computational Systems at Ludwig-Maximilian University of Munich, Germany. He received his B.Sc. degree in Computer and Systems Engineering, 2009, and M.Sc. degree in Electrical Engineering, 2014 from the Faculty of Engineering, Minia University, Egypt. Most of his prior work focused on efficient hardware and software implementations of embedded control systems targeting various computing platforms. In 2016, he jointed the HyConSys lab as a PhD student. His research spans: automated synthesis of correct-by-construction control software for safety-critical systems; formal verification of Cyber-physical systems including embedded control systems, real-time systems, and networked systems; and efficient GPGPU- and HW-based design and implementations of safety-critical control software.

Talk's Recording: View on Youtube

Abstract: This talk will focus on a class of decision making problems in online optimization settings. The literature on online optimization is extremely rich and its connections to many other areas, particularly control theory, has been explored in recent years. Unlike the classical setting of online optimization, where the decisions of the learner are solely chosen according to a cost function, in many realistic scenarios decisions are inputs to a control system. The regret function is defined as the difference between the accumulated costs incurred by control actions made in hindsight using previous states and the cost incurred by the best fixed admissible policy when all cost functions are known in advance. The talk will focus on an online setting of the linear quadratic Gaussian optimal control problem on a sequence of quadratic cost functions. I introduce a modified online Riccati update scheme that under some boundedness assumptions, leads to logarithmic regret bounds, improving the best known bounds in the literature. In relation to the mentioned boundedness, and as a by-product, I will layout some sharp monotonicity contrasts between the classical Riccati difference equation and the less commonly used Newton-Hewer update.

Bio: Bahman Gharesifard is an Associate Professor with the Department of Mathematics and Statistics at Queen's University. He held postdoctoral positions with the Department of Mechanical and Aerospace Engineering at University of California, San Diego 2009-2012 and with the Coordinated Science Laboratory at the University of Illinois at Urbana-Champaign from 2012-2013. He received the 2019 CAIMS-PIMS Early Career Award, jointly awarded by the Canadian Applied & Industrial Math Society and the Pacific Institute for the Mathematical Sciences, a Humboldt research fellowship from the Alexander von Humboldt Foundation in 2019, an NSERC Discovery Accelerator Supplement in 2019, and the SIAG/CST Best SICON Paper Prize 2021. He was a finalist (as an advisor) for the Best Student Paper Award at the American Control Conference in 2017. He received the Engineering and Applied Science First Year Instructor Teaching Award in 2015 and 2017. He serves on the Conference Editorial Board of the IEEE Control Systems Society, and as an Associate Editor for the IEEE Control System Letters and IEEE Transactions on Network Control Systems. His research interests include various topics in control theory including distributed control and optimization, intersections of machine learning and control theory, geometric control theory, social and economic networks, and game theory.

Abstract: In control problems for network systems, we aim to achieve global control objectives through distributed controllers that are limited in their sensing, actuation and connectivity. Ideally, the system’s behavior should also be scalable – its performance should be uniform with respect to the network size. In this talk, I will highlight situations where these objectives cannot be met due to fundamental limitations. We consider networked dynamical systems with linear consensus-like feedback control. Such systems can model, for example, automatic cruise control in vehicle platoons or frequency control in electric power networks. I will show that there are fundamental limitations to localized feedback control that limit the possibility to scale these systems into large networks. This means that commonly proposed control strategies may not allow for long vehicle platoons or, in the context of power networks, a highly distributed power generation paradigm. I will discuss the underlying reasons for these limitations, how they depend on topological and algorithmic properties of the system, and how they may, in certain cases, be alleviated.

Bio: Emma Tegling is a Senior Lecturer with the Department of Automatic Control at Lund University, Sweden. She received her Ph.D. degree in Electrical Engineering from KTH Royal Institute of Technology, Stockholm, Sweden in 2019, and her M.Sc. and B.Sc. degrees, both in Engineering Physics, from the same institute in 2013 and 2011, respectively. Between 2019-2020 she was a Postdoctoral Research Fellow with the Institute of Data, Systems, and Society (IDSS) at the Massachusetts Institute of Technology (MIT), Cambridge, USA. She has also spent time as a visiting researcher at Caltech, the Johns Hopkins University, and UC Santa Barbara. Emma's research interests are within analysis and control of large-scale networked systems. She has a particular interest for control challenges in distributed electric power networks, and lately, the COVID-19 pandemic.