Fall 2015 Seminars
Design of Robotics and Embedded systems, Analysis, and Modeling Seminar (DREAMS)
The Design of Robotics and Embedded systems, Analysis, and Modeling Seminar (DREAMS) occurs weekly on Tuesdays or Mondays from 4.10-5.00 p.m. in 540A/B Cory Hall.
The Design of Robotics and Embedded systems, Analysis, and Modeling Seminar topics are announced to the DREAMS list, which includes the chessworkshop workgroup, which includes the chesslocal workgroup.
Information on the seminar series might be useful for potential speakers. If you have any questions about DREAMS, please contact Dorsa Sadigh. If you want to subscribe to our mailing list, please drop me a line.
Seminars from previous semesters can be found here.
Robust Navigation and Tracking in Dynamic Environments
Sep 08, 2015, 4-5pm, 540 A/B Cory, Songhwai Oh, Seoul National University.
With the recent development in robotics, we can expect that more service robots will be assisting us in the near future in places, such as offices, malls, and homes. But, for a robot to coexist with humans and operate successfully in crowded and dynamic environments, a robot must be able to act safely and harmoniously with human participants in the environment. In this talk, I will describe our recent work on robust navigation using Gaussian process regression and robust target tracking using chance-constrained optimization.
An autoregressive Gaussian process (AR-GP) motion model is developed to predict the future trajectories of pedestrians using measurements from the partially observable egocentric view of a robot. A robot is controlled in real-time based on predicted pedestrian trajectories using the proposed AR-GP motion controller. In order to make the AR-GP motion model robust against outliers and noises, a structured low-rank matrix approximation method using nuclear-norm regularized l1-norm minimization is developed to approximate kernel matrices. A leveraged non-stationary kernel function is proposed to incorporate both positive and negative training samples, in order to speed up the learning process. Using the AR-GP motion model, we have developed a robust target tracking algorithm based on chance-constrained optimization for a mobile sensor with bounded fan-shaped sensing regions, such that the tracking success probability is guaranteed and the travel distance is minimized. I will also describe experimental results from the proposed algorithms.
This talk is based on joint work with Sungjoon Choi, Eunwoo Kim, and Yoonseon Oh.
Songhwai Oh received the B.S. (with highest honors), M.S., and Ph.D. degrees in electrical engineering and computer sciences from the University of California, Berkeley, in 1995, 2003, and 2006, respectively. He is currently an Associate Professor in the Department of Electrical and Computer Engineering, Seoul National University, Seoul, Korea. Before his Ph.D. studies, he was a Senior Software Engineer at Synopsys, Inc. and a Microprocessor Design Engineer at Intel Corporation. In 2007, he was a Postdoctoral Researcher in the Department of Electrical Engineering and Computer Sciences, University of California, Berkeley. From 2007 to 2009, he was an Assistant Professor of electrical engineering and computer science in the School of Engineering, University of California, Merced. His research interests include cyber-physical systems, robotics, computer vision, and machine learning.
Optimal Control with Physical and Cognitive State
Sep 15, 2015, 4-5pm, 540 A/B Cory, Anca Dragan, University of California, Berkeley.
The goal of my research is to enable robots to work with, around, and in support of people, autonomously producing behavior that reasons about both their function and their interaction with humans. I aim to develop a formal understanding of interaction that leads to algorithms which are informed by mathematical models of how humans interact with robots, enabling generalization across robot morphologies and interaction modalities.
In this talk, I will focus on one specific instance of this agenda: autonomously generating motion for coordination during human-robot collaborative manipulation. Most motion in robotics is solely functional: industrial robots move to package parts, vacuuming robots move to suck dust, and personal robots move to clean up a dirty table. This type of motion is ideal when the robot is performing a task in isolation. Collaboration, however, does not happen in isolation, and demands that we move beyond solely functional motion. In collaboration, the robot's motion has an observer, watching and interpreting the motion Ã¢ï¿½ï¿½ inferring the robot's intent from the motion, and anticipating the robot's motion based on its intent.
My work integrates a mathematical model of these inferences into motion planning, so that the robot can generate motion that matches people's expectations and clearly conveys its intent. In doing so, I draw on action interpretation theory, Bayesian inference, constrained trajectory optimization, and interactive learning. The resulting motion not only leads to more efficient collaboration, but also increases the fluency of the interaction as defined through both objective and subjective measures.
Anca Dragan is a new Assistant Professor in UC Berkeley's EECS Department. She completed her PhD in Robotics at Carnegie Mellon. She was born in Romania and received her B.Sc. in Computer Science from Jacobs University in Germany in 2009. Her research lies at the intersection of robotics, machine learning, and human-computer interaction: she works on algorithms that enable robots to seamlessly work with, around, and in support of people. Anca's research and her outreach activities with children have been recognized by the Intel Fellowship and by scholarships from Siebel, the Dan David Prize, and Google Anita Borg.
A Temporal Logic Approach to Information-Flow Control
Sep 22, 2015, 4-5pm, 540 A/B Cory, Markus Rabe, University of California, Berkeley.
Information-flow control is a principled approach to prevent vulnerabilities in programs and other technical systems. In information-flow control we define information-flow properties, which are sufficient conditions for when the system is secure in a particular attack scenario, and then enforce that the given system adheres to its properties. Information-flow properties refer to how the information flows through the system and standard verification approaches seemed to be inapplicable.
In this talk I present recent work on a temporal logic approach to information-flow control. We discuss how temporal logics can be used to provide a simple formal basis for the specification and enforcement (in our case model checking) of information-flow properties. The main challenge of this approach is that the standard temporal logics are unable to express these properties. They lack the ability to relate multiple executions of a system, which is the essential feature of information-flow properties. We thus extend temporal logics by the ability to quantify over multiple executions and to relate them using boolean and temporal operators.
We show that the approach supports a wide range of previously known and also new information-flow properties with a single model checking algorithm. We demonstrate the effectiveness of the approach along case studies in hardware security.
Markus Rabe is a new postdoc in Sanjit Seshia's group in UC Berkeley's EECS Department. He spent his PhD studies at Saarland University under the supervision of Bernd Finkbeiner. His research interests include security, formal verification, and synthesis.
Formal Methods for Dynamical Systems
Sep 28, 2015, 4-5pm, 540 A/B Cory, Calin Belta, Boston University.
In control theory, complex models of physical processes, such as systems of differential equations, are usually checked against simple specifications, such as stability and set invariance. In formal methods, rich specifications, such as languages and formulae of temporal logics, are checked against simple models of software programs and digital circuits, such as finite transition graphs. With the development and integration of cyber physical and safety critical systems, there is an increasing need for computational tools for verification and control of complex systems from rich, temporal logic specifications. The formal verification and synthesis problems have been shown to be undecidable even for very simple classes of infinite-space continuous and hybrid systems. However, provably correct but conservative approaches, in which the satisfaction of a property by a dynamical system is implied by the satisfaction of the property by a finite over-approximation (abstraction) of the system, have received a lot of attention in recent years. The focus of this talk is on discrete-time linear systems, for which it is shown that finite abstractions can be constructed through polyhedral operations only. By using techniques from model checking and automata games, this allows for verification and control from specifications given as Linear Temporal Logic (LTL) formulae over linear predicates in the state variables. The usefulness of these computational tools is illustrated with various examples.
Calin Belta is a Professor in the Department of Mechanical Engineering, Department of Electrical and Computer Engineering, and the Division of Systems Engineering at Boston University, where he is also affiliated with the Center for Information and Systems Engineering (CISE) and the Bioinformatics Program. His research focuses on dynamics and control theory, with particular emphasis on hybrid and cyber-physical systems, formal synthesis and verification, and applications in robotics and systems biology. Calin Belta is a Senior Member of the IEEE and an Associate Editor for the SIAM Journal on Control and Optimization (SICON) and the IEEE Transactions on Automatic Control. He received the Air Force Office of Scientific Research Young Investigator Award and the National Science Foundation CAREER Award.
Bridging the Gap between Research and Practice: Formal Methods in the Automotive Domain
Oct 05, 2015, 4-5pm, 540 A/B Cory, Jyotirmoy Deshmukh, Toyota Technical Center.
At the heart of an automobile are its engine and powertrain, and the operation of these components is controlled by embedded software on electronic control units (ECUs). The paradigm of model-based development (MBD) has become the de facto standard for designing such control software. MBD designs of control software range from feature-level models to application-level and even entire system-level models. On the other hand, models of the plant (e.g. the engine), can range from simple physics-based models to high-fidelity models incorporating test-data. The advantage of MBD is in its ability to design, validate, and analyze the closed-loop model of the plant and the controller, often well before the actual hardware components become available. Unfortunately, even the simplest closed-loop model of an automotive powertrain subsystem is a complex cyber-physical system with highly nonlinear and hybrid dynamics, and reasoning about the correctness of such closed-loop models is a formidable task. In this talk, we introduce two challenges in reasoning about industrial-scale closed-loop control models: (1) scaling verification or bug-finding techniques to engine control software, and (2) formalisms to express correctness and performance requirements for such models. We survey some of the existing work done to address such questions, and present some promising directions for future work.
Jyotirmoy V. Deshmukh is a senior engineer at the Toyota Technical Center in Gardena, CA, where he does research on verification and validation techniques for powertrain control software. Before joining Toyota, Jyotirmoy was a post-doctoral researcher at the University of Pennsylvania. He received his Ph.D. from the University of Texas at Austin, and his dissertation focused on verification of concurrent and sequential software libraries. JyotirmoyÃ¢ï¿½ï¿½s research interests broadly include: verification and correct-by-construction design of cyberphysical systems, hybrid systems theory, automata theory, concurrency, temporal logics and formal software verification techniques.
A Theory of Privacy for Cyber-Physical Systems with Applications in Energy Systems
Oct 13, 2015, 4-5pm, 289 Cory, Shuo Han, University of Pennsylvania.
As many cyber-physical systems start to rely on collecting user data for more efficient operation, privacy has emerged as a concern among participating users. In this talk, I will discuss two frameworks that formalize the notion of privacy (differential privacy and information-theoretic privacy) from a unified point of view based on detection theory. I will demonstrate the applications of the two frameworks in energy systems through two case studies. (i) Private distributed charging of electric vehicles: It has been shown that the (non-private) distributed charging problem can be solved using distributed gradient descent. However, the messages exchanged between the center mediator and users may be exploited to breach the privacy of users. We show that differential privacy can be preserved by introducing additive noise to the gradients. We also quantify the trade-off between the level of privacy and the loss of utility using tools from optimization theory. (ii) Private smart metering with internal energy storage: We propose a new information-theoretic metric of privacy in order to handle the privacy of events (e.g., energy usage within any given time slot). The new metric is used to analyze the privacy of a smart metering system that uses internal energy storage as a buffer to hide distinctive energy usage patterns. The results quantify how the amount of energy storage helps improve the level of privacy.
Shuo Han is a postdoctoral researcher in the Department of Electrical and Systems Engineering at the University of Pennsylvania. He received his B.E. and M.E. in Electronic Engineering from Tsinghua University in 2003 and 2006, and his Ph.D. in Electrical Engineering from the California Institute of Technology in 2013. His current research focuses on developing rigorous frameworks for data-driven decision making that enable reliable and efficient operations of networked cyber-physical systems, including many smart city applications such as power and transportation networks. He was a finalist for the Best Student Paper Award at the 2013 American Control Conference.
An Interactive Approach to Mobile App Verification
Oct 19, 2015, 4-5pm, 540 A/B Cory, Osbert Bastani, Stanford University.
Static information flow analysis has the potential to greatly aid human auditors in finding privacy violations and malicious software. However, challenges in scaling sound, precise static analyses mean that an auditor must settle for either imprecise results (generating many false positives, which can be time-consuming to discharge) or unsound results (which makes it easy for adversaries to hide malicious behaviors).
We propose a verification approach that relies on interaction with a human auditor to eliminate false positives. In our approach, the static analysis queries the auditor about program properties, and uses abductive inference to minimize the number of queries needed. We apply this approach to solve two problems we have encountered in practice: (i) synthesizing models for relevant library functions, and (ii) removing dead code that causes imprecision in the analysis.
Osbert is a fourth-year Ph.D. student working with Prof. Alex Aiken. He is interested in applying inference techniques to improve program analysis. He currently works on STAMP, which uses program analysis to help a human auditor identify Android malware. He has also worked on applying ideas from program analysis to improve the quality of deep neural nets and decision forests, on Google's static analysis infrastructure, and on interactive recommender systems.
Analysis and software synthesis of KPN applications
Oct 22, 2015, 4-5pm, 400 Cory, Jeronimo Castrillon, TU Dresden.
Programming models based on dataflow or process networks are a good match for streaming applications, common in the signal processing, multimedia and automotive domains. In such models, parallelism is expressed explicitly which makes them well-suited for programming parallel machines. Since todayÃ¢ï¿½ï¿½s applications are no longer static, expressive programming models are needed, such as those based on Kahn Process Networks (KPNs). In these models, tasks cannot be handled as black boxes, but have to be analyzed, profiled and traced to characterize their behavior. This is especially important in the case of heterogenous platforms with many processors of multiple different types. This presentation describes a tool flow to handle KPN applications and gives insights into mapping algorithms for heterogeneous platforms.
Jeronimo Castrillon is a professor in the Department of Computer Science at the TU Dresden, where he is also affiliated with the Center for Advancing Electronics Dresden (CfAED). He received the Electronics Engineering degree with honors from the Pontificia Bolivariana University in Colombia in 2004, the master degree from the ALaRI Institute in Switzerland in 2006 and the Ph.D. degree (Dr.-Ing.) with honors from the RWTH Aachen University in Germany in 2013. His research interests lie on methodologies, languages, tools and algorithms for programming complex computing systems.
Functional Reactive Programming for Real-Time and Cyber-Physical Systems:
Oct 23, 2015, 3-4pm, 540 A/B Cory, Albert Cheng, University of Houston, Texas.
The use of sophisticated digital systems to control complex physical components in real-time has grown at a rapid pace. These applications range from traditional stand-alone systems to highly-networked cyber-physical systems (CPS's), spanning a diverse array of software architectures and control models. Examples include automobile adaptive braking, industrial robotic assembly, medical pacemakers, autonomous (ground, air, and sea) vehicular travel, remote surgery, physical manipulation of nano-structures, search-and-rescue, and space exploration. Since all these applications interact directly with the physical world and often have humans in the loop, we must ensure their physical safety.
Obviously, the correctness of these embedded systems and CPS's depends not only on the effects or results they produce, but also on the time at which these results are produced. For example, when the driver of a car applies the brake, the anti-lock braking controller analyzes the environment in which the controller is embedded (car speed, road surface, direction of travel) and activates the brake with the appropriate frequency within fractions of a second. Both the result (brake activation) and the time at which the result is produced are important in ensuring the safety of the car, its driver and passengers. In a CPS consisting of a multitude of vehicles and communication components with the goal to avoid collisions and reduce traffic congestions, formal safety verification and response time analysis are essential to the certification and use of such systems.
The benefits of using the functional (reactive) programming (FRP) over the imperative programming style found in languages such as C/C++ and Java for implementing embedded and real-time software are several. The functional programming paradigm allows the programmer to intuitively describe safety-critical behaviors of the system, thus lowering the chance of introducing bugs in the design phase. Its stateless nature of execution does not require the use of synchronization primitives like mutexes and semaphores, thus reducing the complexity in programming. However, accurate response time analysis of FRP-based controllers remains a largely unexplored problem. This talk will introduce a framework for accurate response time analysis, scheduling, and verification of embedded controllers implemented as FRP programs.
*Supported in part by the US National Science Foundation Awards No. 1219082 and No. 0720856.
Albert M. K. Cheng is Professor and former interim Associate Chair of the Computer Science Department at the University of Houston (UH), Texas. His research interests center on the design, specification, modeling, scheduling, and formal verification of real-time, embedded, and cyber-physical systems, green/power/thermal-aware computing, software engineering, knowledge-based systems, and networking. He is the founding Director of the UH Real-Time Systems Laboratory. He received the B.A. with Highest Honors in Computer Science, graduating Phi Beta Kappa at age 19, the M.S. in Computer Science with a minor in Electrical Engineering at age 21, and the Ph.D. in Computer Science at age 25, all from The University of Texas at Austin, where he held a GTE Foundation Doctoral Fellowship. He has served as a technical consultant for a number of organizations, including IBM and Shell, and was also a Visiting Professor in the Departments of Computer Science at Rice University and at the City University of Hong Kong. He is a co-founder of ZapThru.com, where he is currently the Chief Strategy and Technology Director.
Dr. Cheng is the author/co-author of over 200 refereed publications in leading journals (including IEEE Transactions on Computers, IEEE Transactions on Software Engineering, and IEEE Transactions on Knowledge and Data Engineering) and top-tier conferences (such as RTSS, RTAS, ICPADS, and ISLPED), and has received numerous awards, including the U.S. National Science Foundation Research Initiation Award and the Texas Advanced Research Program Grant. He has been invited to present seminars, tutorials, panel positions, and keynotes at nearly 100 conferences, universities, and organizations. He is and has been on the technical program committees (including many program chair positions) of over 240 conferences, symposia, workshops, and editorial boards (including the IEEE Transactions on Software Engineering 1998-2003 and the IEEE Transactions on Computers 2011-present).
He has been the Guest Co-Editor of a 2013 Special Issue on Rigorous Modeling and Analysis of Cyber-Physical Systems of the IEEE Embedded Systems Letters, and is the Guest Editor of a 2014 Special Issue on Cyber-Physical Systems of SENSORS. Recently, he has been the Program Co-Chair of the 2013 IEEE International Conference on Service Oriented Computing and Applications (SOCA) and the Program Co-Chair of the System, Models and Algorithms Track of the 2014 IEEE International Conference on Embedded Software and Systems (ICESS) where he delivered an award-winning Keynote on Next-Generation Embedded Systems. Currently, Dr. Cheng is the Program Chair of the First Workshop on Declarative Programming for Real-Time and Cyber-Physical Systems (DPRTCPS) at the IEEE Real-Time Systems Symposium (RTSS), San Antonio, Texas, USA, December 1, 2015; and the Program Chair of the International Symposium on Software Engineering and Applications (SEA), Marina del Rey, California, USA, October 26-28, 2015.
Dr. Cheng is the author of the popular senior/graduate-level textbook entitled Real-Time Systems: Scheduling, Analysis, and Verification (John Wiley & Sons), 2nd printing with updates, 2005. He is a Senior Member of the IEEE; an Honorary Member of the Institute for Systems and Technologies of Information, Control and Communication (INSTICC); and a Fellow of the Institute of Physics (IOP). He is the recipient of the 2015 UH Lifetime Faculty Award for Mentoring Undergraduate Research.
Oct 26, 2015, 4-5pm, 540 A/B, Dejan Milutinovic, University of California, Santa Cruz.
While stochastic processes are useful models for drifts influencing the motion of nonholonomic vehicles, they can also be used to model a family of possible trajectories in single- or multi-vehicle systems. This results into nonlinear stochastic dynamical models of these systems. Since nonlinear control problems can be difficult enough in themselves, it seems that this approach adds yet another layer of complexity. Interestingly enough, although stochastic dynamical systems are complex, they have properties that allow us to solve control problems computationally easier than in the case of their deterministic counterparts. These solutions are attractive because computed controls anticipate stochastic process uncertainties and balance the intensity of control actions with respect to the intensity of stochastic processes. Moreover, they show the potential to solve control problems with many degrees of freedom. This talk reviews my research with several examples, including single and multiple vehicles, micro-robots and cell biology applications, scaled-down robot experiments for airport ramp area aircraft maneuvers and an analysis of human arm motions.
Dejan Milutinovic earned Dipl.-Ing (1995) and Magister's (1999) degrees in electrical engineering from the University of Belgrade, Serbia and a doctoral degree in electrical and computer engineering (2004) from Instituto Superior Tecnico, Lisbon, Portugal.
From 1995 to 2000 he worked as a research engineer in the Automation and Control Division of Mihajlo Pupin Institute, Belgrade, Serbia. His doctoral thesis was the first runner-up for the best PhD thesis of European Robotics in 2004 by EURON. He won the NRC award of the US Academies in 2008 and Hellman Fellowship in 2012.
Dr. Milutinovic is currently an associate professor in the Department of Computer Engineering, UC Santa Cruz. His research interests are in the area of modeling and control of stochastic dynamical systems applied to robotics. He is an associate editor of the ASME Journal of Dynamic Systems, Measurement, and Control.
Programming Languages for High-Assurance Air Vehicles
Nov 02, 2015, 4-5pm, 540 A/B Cory, Lee Pike, Galois INC.
We describe the use of embedded domain-specific languages to improve programmer productivity and increase software assurance in the context of building a fully-featured autopilot for unpiloted aircraft. This work is performed for DARPA under the High-Assurance Cyber Military System (HACMS) program.
Lee Pike manages the Cyber-Physical Systems program at Galois, Inc., a company specializing in software-intensive critical systems. He has been the PI on approximately $10 million of R&D projects funded by NASA, DARPA, AFRL, and other federal agencies. His research focuses on applying techniques from functional programming, run-time verification, and formal verification to the areas of operating systems, compilers, cryptographic systems, avionics, and control systems. Previously, he was a research scientist in the NASA Langley Formal Methods Group and has a Ph.D in Computer Science from Indiana University.
Hardware Security: An Emerging Threat Landscape and Possible Solutions
Nov 10, 2015, 4-5pm, 400 Cory, Siddharth Garg, NYU Polytechnic School of Engineering.
For economic reasons, the design and fabrication of semiconductor ICs is increasingly outsourced. This comes at the expense of trust. An untrusted entity, for instance a malicious designer or semiconductor foundry could pirate the design, or worse, maliciously modify it to leak secret information from the chip or sabotage its functionality.
In this talk, I will present my recent work on two defense mechanisms to secure computer hardware against such attacks. The first is split manufacturing, which enables a designer to partition a design across multiple chips, fabricate each separately, and "glue" them together after fabrication. Since each foundry only sees a part of the design, its ability to infer the design intent is hindered. I will propose a quantitative notion of security for split manufacturing and explore the resulting cost-security trade-offs.
In the second part of the talk, I will discuss another defense mechanism: logic obfuscation. Previous work has proposed logic obfuscation with seemingly strong security guarantees. I will present a new and effective attack on all existing logic obfuscation techniques, and the security implications of this new attack.
Siddharth Garg received his Ph.D. degree in Electrical and Computer Engineering from Carnegie Mellon University in 2009, and a B.Tech. degree in Electrical Engineering from the Indian Institute of Technology Madras. He joined NYU in Fall 2014 as an Assistant Professor, and prior to that, was an Assistant Professor at the University of Waterloo from 2010-2014. His general research interests are in computer engineering, and more particularly in secure, reliable and energy-efficient computing. For his research, Siddharth has received best paper awards at the USENIX Security Symposium 2013, at the Semiconductor Research Consortium (SRC) TECHCON in 2010, and the International Symposium on Quality in Electronic Design (ISQED) in 2009. Siddharth also received the Angel G. Jordan Award from ECE department of Carnegie Mellon University for outstanding thesis contributions and service to the community.
Inverse Optimization with Noisy Data
Nov 23, 2015, 4-5pm, 540 A/B Cory, Anil Aswani, University of California, Berkeley.
Motivated by predictive modeling problems in personalized chronic disease management, we present a framework for solving inverse optimization problems with noisy data. These are problems in which noisy measurements of minimizers to a parametric optimization problem are observed and then used to estimate the unknown parameters of the optimization problem. Though a number of approaches have been developed, these approaches require no noise in the measured data and no modeling mismatch. In this talk, we present our framework that deals with the case where the parametric optimization problem is convex, and show this framework is risk consistent (asymptotically provides best possible predictions) or estimation consistent (asymptotically estimates true parameters) under appropriate conditions. Numerically, we provide three optimization formulations to solve inverse problems using our framework, including a new approach to solving bilevel optimization problems, a new approximation algorithm, and an approach based on integer programming. We conclude with a case study on applying these approaches to real data involving personalized chronic disease management.
Anil Aswani is an assistant professor in Industrial Engineering and Operations Research (IEOR) at the University of California, at Berkeley. He received a BS in Electrical Engineering in 2005 from the University of Michigan, in Ann Arbor, and a PhD in Electrical Engineering and Computer Sciences with Designated Emphasis in Computational and Genomic Biology in 2010 from the University of California, at Berkeley. His research focuses on designing new statistical and optimization methods that utilize big data in order to generate empirical models of human-behavior within complex systems, which can then be used for better understanding, optimization, configuration, and design of these systems.
Constructive Feedback Control Design for Hybrid Systems
Nov 30, 2015, 4-5pm, 540 A/B Cory, Ricardo Sanfelice, University of California, Santa Cruz.
A mathematical framework to model hybrid dynamical systems with inputs and outputs will be presented. Motivated by control design challenges in robotics and power system applications, tools for the design of feedback controllers will be introduced. The focus will be on tools guaranteeing asymptotic stability of the hybrid closed-loop system with robustness to a general class of perturbations. In particular, relaxed Lyapunov sufficient conditions for stability and attractivity, as well as control Lyapunov functions (CLFs) for the design of feedback laws will be presented. It will be shown that the proposed CLF-based feedback law provides an estimate of the upper value function of the zero-sum hybrid dynamical game, where one player assigns the control and another the disturbance. The tools will be exercised in robotic and power systems applications.
Ricardo G. Sanfelice is an Associate Professor at the Department of Computer Engineering at University of California, Santa Cruz. He received his M.S. and Ph.D. degrees from UCSB in 2004 and 2007, respectively. He held a Postdoctoral Associate position at MIT during 2007 and 2008. Prof. Sanfelice is the recipient of the 2013 SIAM Control and Systems Theory Prize, the NSF CAREER award, the Air Force Young Investigator Research Award (YIP), the 2010 IEEE Control Systems Magazine Outstanding Paper Award, and the 2012 STAR Higher Education Award for his contributions to STEM education. He is Associate Editor for Automatica. His research interests are in modeling, stability, robust control, observer design, and simulation of nonlinear and hybrid systems with applications to robotics, power systems, aerospace, and biology.
|You are not logged in|
|©2002-2018 U.C. Regents|