The technical program will be available in early February, 2020.
Abstract: Since the advent of Shannon's paradigm, it has been ubiquitously recognized that, under the perfect band-limiting hypothesis, we can reconstruct and control any signals below the so-called Nyquist frequency, but also nothing beyond. It appears as if this were an indisputable limit for digital processing, and people have almost blindly accepted this limitation, without examining the basic assumptions. A moment of a careful examination would immediately reveal the fact that the perfect band-limiting hypothesis is merely a sufficient condition to allow for a perfect signal recovery, and there can be indeed several ways to go in a different direction. Through the study of sampled-data control systems, the present author and collaborators found that there is a great potential of this new theory to be applied to signal processing if one removes the hard requirement of perfect reconstruction and adopt some mild frequency model. The new methodology, now commonly known as the “YY Filter”, has made a spectacular success, allowing reconstruction beyond the Nyquist frequency, and also led to commercial success. This talk gives an overview of this discovery and the development in the past two decades, and show how we have succeeded in making it to be a commercial success. We will conclude the talk by reviewing our recent approach toward precise tracking/rejection of signals beyond the Nyquist frequency. This last topic is much needed in modern technology.
Biography: Yutaka Yamamoto obtained his Ph. D degree from the university of Florida in1978, under the supervision of Professor Rudolf Kalman. He then obtained a faculty position in Kyoto University, and stayed there until his retirement in 2015. He is Professor Emeritus of Kyoto University. His research interest includes linear system theory, infinite-dimensional systems, their algebraization, sampled-data systems, and digital signal processing. His current interest is in control of signals beyond the Nyquist frequency. He is a winner of several awards, including the G. S. Axelby outstanding paper award, Tateishi Prize, Commendation of Science and Technology of the minister of Education and Science, Transition to Practice Award of the CSS, and others. He is a fellow of the IEEE, IFAC and SICE. He was President of the IEEE CSS, and ISCIE, and served as the General Chair of the MTNS 2006. He is also an honorary member of SICE and ISCIE, as well as a life fellow of the IEEE.
Abstract: A linear time-invariant system is said to be (internally) positive if its state and output are nonnegative for any nonnegative initial state and any nonnegative input. In the former part of this talk, we focus on the characterization of the weighted L1-induced norm of positive systems in terms of linear programming problems (LPs). As the main result, we show that the weighted L1-induced norm plays a fundamental role for the stability analysis of interconnected positive systems. On the other hand, in the latter part of this talk, we consider the Hankel-type Lq/Lp induced norms across a single switching over two positive systems. The norms are defined as the induced norms from vector-valued Lp-past inputs to vector valued Lq-future outputs across switching at the time instant zero. Due to the strong positivity property, we can explicitly characterize the Hankel-type Lq/Lp induced norms even for p, q being 1, 2, ∞, and these are useful to evaluate the performance deterioration caused by switching quantitatively. In particular, we will show that some of them are given in the form of LP and this enables us to synthesize the induced-norm-optimal-parameters by solving geometric programming problems (GPs). We finally illustrate the effectiveness of the GP-based-synthesis method by applying it to the optimal parameter tuning problem of the Foschini-Miljanic algorithm for power control in wireless network communication.
Biography: Yoshio Ebihara received the B.E., M.E., and D.E. degrees in electrical engineering from Kyoto University, Kyoto, Japan, in 1997, 1999, and 2002, respectively. In 2002, he joined the Department of Electrical Engineering, Kyoto University, where he held positions as an Assistant Professor, Lecturer, and Associate Professor. Since 2019, he has been a full professor in the Graduate School and Faculty of Information Science and Electrical Engineering, Kyushu University, Japan. In 2010, he held a Visiting Researcher position at LAAS-CNRS Toulouse, France. His research interests include convex optimization in control and positive system analysis and synthesis. Prof. Ebihara is currently an Associate Editor of Automatica, and the chair of IEEE CSS Technical Committee on Robust and Complex Systems. He is also the General Chair of the 10th IFAC Symposium on Robust Control Design, to be held in Kyoto, Japan, in 2021.
Abstract: Recent radical evolution in distributed sensing, computation, communication, and actuation has fostered the emergence of cyber-physical network systems. Examples cut across a broad spectrum of engineering and societal fields. Regardless of the specific application, one central goal is to shape the network collective behavior through the design of admissible local decision-making algorithms. This is nontrivial due to various challenges such as the local connectivity, imperfect communication, model and environment uncertainty, and the complex intertwined physics and human interactions. In this talk, I will present our recent progress in formally advancing the systematic design of distributed coordination in network systems. We investigate the fundamental performance limit placed by these various challenges, design fast, efficient, and scalable algorithms to achieve (or approximate) the performance limits, and test and implement the algorithms on real-world applications.
Biography: Na Li is a Thomas D. Cabot associate professor in Electrical Engineering and Applied Mathematics of the J. Paulson School of Engineering and Applied Sciences at Harvard University. She received her Bachelor degree in Mathematics from Zhejiang University in 2007 and Ph.D. degree in Control and Dynamical systems from California Institute of Technology in 2013. She was a postdoctoral associate of the Laboratory for Information and Decision Systems at Massachusetts Institute of Technology 2013-2014. She has joined Harvard University since 2014. Her research lies in distributed learning, optimization, and control of cyber-physical networked systems. She received NSF career award (2016), AFSOR Young Investigator Award (2017), Harvard PSE Accelerator Award (2017), ONR Young Investigator Award (2019), Donald P. Eckman Award (2019), and some paper awards.
Abstract: Worldwide, there is significant growth in the installed capacity of wind energy based power plants. Whilst they provide a reprieve from greenhouse gas emissions, the inherent stochasticity in the availability of wind introduces problems of deviations from forecasts. Sans efficient and inexpensive solutions for grid-level energy storage, wind power plants have to rely on economically unviable operational methods such as curtailment. The growing presence of plug-in electric vehicle (PEV) fleets in electricity distribution systems may provide an avenue for avoiding curtailments by scheduling their charging during over-supply of wind energy. In this talk, we will explore a data-driven methodology for studying existing pricing mechanisms in a North American metropolis, located in a wind energy-rich grid, with an increased adoption rate of PEVs. Our study shows that the existing pricing methods for encouraging PEV charging to consume excess wind are sub-optimal. We will justify a need for an alternative dynamic pricing mechanism - Time of Use (ToU) - dedicated to PEVs for maximizing the usage of available energy from wind. Such a dedicated ToU can help avoid high costs for the supplier and deliver benefits to the responsive consumers. Using real data and a machine learning methodology for data decomposition, we have designed a ToU dedicated for PEVs in the city under consideration. We will conclude this talk with some thoughts on the future directions of this research.
Biography: Sid Suryanarayanan is from Chennai, India. He received the M.S. and the Ph.D. in electrical engineering in 2001 and 2004, respectively, from Arizona State University. Currently, he is a professor in the Dept. of ECE at Colorado State University. He has over 100 technical publications in journals and conferences. Suryanarayanan is the co-editor of two books in electric power engineering. His recent accolades include a 2018 R&D100 Award, the 2017 IEEE Eta-Kappa-Nu (HKN) C. Holmes MacDonald Outstanding Teaching Award and the 2015 IEEE Power and Energy Society (PES) Outstanding Young Engineer Award.
The information on the banquet will be announced soon.