Validated Numerics - a short introduction to rigorous computations

Wednesday January 16, 13:15 (1:15pm), Wigforssalen, Halmstad University

Click here for a video recording of the lecture.

We will present an efficient means of performing numerical computations with rigorous error bounds. The basic idea is to use set-valued mathematics as the underlying framework. This enables us to change focus from approximating the solution to enclosing the same. We will apply these techniques to several problems, ranging from simple root-finding and quadrature to parameter estimation.

About Professor Warwick Tucker

After receiving his doctoral degree in mathematics at Uppsala University in 1998, proving that the Lorenz attractor exists, Tucker spent two years at IMPA (Rio de Janeiro, Brasil) as a postdoctoral fellow. During the years 2000 to 2002, Tucker held a H.C. Wang assistant professorship at Cornell University (Ithaca, USA) teaching and doing research in mathematics. During this period, Tucker was awarded the Swedish Mathematical Society's Wallenberg Prize, and the R.E. Moore Prize for Applications of Interval Analysis. Upon his return to Sweden, Tucker was awarded a five-year research fellowship from the Swedish Royal Academy of Sciences. In 2004, Tucker was awarded the European Mathematical Society's Prize for distinguished contributions in Mathematics. In 2007, Tucker formed the CAPA group at the University of Bergen (Bergen, Norway). In 2009, the group moved to its current location at Uppsala University. Tucker was promoted to full professor in 2011.

Claus Führer on Differential-Algebraic Equations in Multibody Dynamics

The event happens on a Friday.  A previous announcement had the wrong day.  Apologies for any confusion this may have caused.

Friday November 23, 13:15 (1:15pm), Wigforssalen, Halmstad University

Click here for a video recording of the lecture.

Equations of motion of constrained multibody systems are differential-algebraic equations of index 3. In this talk an introduction to this class of ordinary differential equations is given and the concept of index and index reduction is introduced. Special focus will be put on the numerical treatment of these equations by multistep methods. Equations of motion can also be transformed to state space form (minimal coordinate formulation), there is a formulation based on coordinate partitioning and they occur as overdetermined differential algebraic systems. These variants have an direct impact on the numerical solution process and its quality. This will be the topic of the last part of the talk.

About Professor Claus Führer

Claus Führer is a Professor of Scientific Computing at Lund University (LTH) and is specialized on simulation of mechanical systems. Before he was at DLR (German Center for Aerospace and Aeronautics) and there responsible for the numerical kernel of SIMPACK. He published a monograph on Numerical Methods in Multibody Dynamics and acted as consultant for various companies.

Karl-Erik Årzén on Simulation of Cyber-Physical Control Systems

Wednesday November 14, 13:15 (1:15pm), Haldasalen, Halmstad University

Click here for a video recording of the lecture.

Whether a control application should be considered CPS or not depends on (at least) three different things. A control system is a CPS when the temporal effects of the implementation platform caused by computing and communication, needs to modeled and included in the design at a more detailed levels than what is traditionally done in computer-based control (periodic sampling and constant latencies). Second, control applications of a CPS nature are typically more distributed and decentralized in nature than the classical, more centralized, control approaches. Third, a control application can be considered to be CPS when the system under control itself is a computing and/or communication system, e.g., a data center on an embedded MPSoC. Simulation is a technique that is well-established in the control community as a complement to more formal verification techniques. Correct simulation of cyber-physical control systems requires support for co-simulation of the temporal effects of real-time kernels and networks together with the continuous-time dynamics of the physical system under control. This is supported by the TrueTime simulator. TrueTime has been available since around 2000 in a Matlab/Simulink version. Currently TrueTime is being ported to Modelica. The equation-based and object-oriented nature of Modelica makes it widely superior to Simulink for representing the Physical part of a CPS system. The new synchronous support for hybrid and discrete-time systems currently being developed within the Modelica community, also makes it ideally suited for representing discrete-time controllers. With the additional support provided by TrueTime for simulating real-time kernels and networks Modelica is also able to handle the Cyber part of a CPS system. The talk will focus on the new features in TrueTime, e.g., support for multicore kernels and bandwidth reservations and the Modelica version of TrueTime currently being developed jointly by Lund University and Vanderbilt University. The version is based on the open model exchange interface FMI (Flexible Mock-Up Interface), an open source alternative to Simulink's S-functions.

About Professor Karl-Erik Årzén

Karl-Erik Årzén a Professor of Automatic Control at Lund University, Lund, Sweden. He received his PhD in 1987 from the same university and since then also worked for ABB Corporate Research. Arzen's research lies in the intersection of control engineering and embedded real-time systems, with a special emphasis on co-design of control and computing systems, and on feedback computing for embedded systems. Arzen has been strongly involved in the European Networks of Excellence Artist2 and ArtistDesign on embedded system design as well as participated in several EU STREP and IP projects. He is currently the co-director of the strategic ICT research area ELLIIT.

Charles Consel on Design-driven Development of Dependable Applications

Tuesday June 19, 13:15 (1:15pm), Wigforssalen, Halmstad University

Click here for a video recording of the lecture.

Making an application dependable demands that its functional and non-functional requirements be stringently fulfilled throughout its development process. In this context, a design-driven development approach has the key advantage of enabling requirements to be traced from their high-level design forms to the resulting executable artifact. However, because such approaches are mostly general purpose, they provide little design guidance, if any. This situation makes unpredictable the coherence and the conformance of an application with respect to its requirements.

To address this situation, we propose an approach that leverages a design-driven development process dedicated to a specific paradigm. This approach guides the verification of the coherence and conformance of an application throughout its development. We demonstrate the benefits of our approach by applying it to a realistic
case study in the avionics domain.

About Professor Charles Consel

Charles Consel is a Professor of Computer Science at University of Bordeaux I. He served on the faculty of Yale University, Oregon Graduate Institute and the University of Rennes.

He leads the Phoenix group at INRIA. He has been designing and implementing Domain-Specific Languages (DSLs) for a variety of areas including device drivers, programmable routers, stream processing, and telephony services. These DSLs have been validated with real-sized applications and showed measureable benefits compared to applications written in general-purpose languages.

His research contributions cover programming languages, software engineering, operating systems, pervasive computing, and assisted living.

Aaron D. Ames on Human-Inspired Control of Bipedal Robots

Tuesday June 5th 2012, 13:15 (1:15pm), Wigforssalen, Halmstad University

Click here for a video recording of the lecture.

Bipedal robots provide an important example of a cyber-physical system (CPS). As a result, understanding the process of realizing formal results experimentally, and the software structures needed to do so, can yield important insights into software synthesis for complex CPS. This talk presents the process of formally achieving human-like bipedal robotic walking by synthesizing controllers inspired by human locomotion data, discusses the software structures used in realizing these controllers and demonstrates these methods through experimental realization on two bipedal robots: AMBER and NAO.

Motivated by the hierarchical control present in humans, the fundamental principle behind this process is that the essential information needed to understand walking is encoded by a simple class of functions canonical to human walking. In other words, we view the human as a complex system, or "black box," and outputs of this system (as computed from human locomotion data) are presented that appear to characterize its behavior—thus yielding low dimensional characterization of human walking. By considering the equivalent outputs for the bipedal robot, a nonlinear controller can be constructed that drives the outputs of the robot to the output of the human; moreover, the parameters of this controller can be optimized so that stable robotic walking is provably achieved while simultaneously producing outputs of the robot that are as close as possible to those of a human. The end result of this process is the automatic generation of bipedal robotic walking that is surprisingly human-like. Extensions of these ideas to different walking behaviors, e.g., stair climbing, will be discussed, along with their application to simulating and designing controllers for prosthetic devices. Finally, the experimental realization of these formal results on two robotic platforms—an underactuated 2D biped, AMBER, and a fully actuated 3D robot, NAO—will be demonstrated.

About Dr. Aaron D. Ames

Dr. Aaron D. Ames is an Assistant Professor in Mechanical Engineering at Texas A&M University, with a joint appointment in Electrical and Computer Engineering. His research interests center on robotics, nonlinear control, and hybrid and cyber-physical systems, with special emphasis on bipedal robots, behavior unique to hybrid systems such as Zeno behavior, and the mathematical foundations of hybrid systems.  Dr. Ames received a BS in Mechanical Engineering and a BA in Mathematics from the University of St. Thomas in 2001, and he received a MA in Mathematics and a PhD in Electrical Engineering and Computer Sciences from UC Berkeley in 2006. At UC Berkeley, he was the recipient of the 2005 Leon O. Chua Award for achievement in nonlinear science and the 2006 Bernard Friedman Memorial Prize in Applied Mathematics. Dr. Ames served as a Postdoctoral Scholar in the Control and Dynamical System Department at the California Institute of Technology from 2006 to 2008.  In 2010 he received the NSF CAREER award for his research on bipedal robotic walking and its applications to prosthetic devices.  

Steven Shladover on Progress Toward Automated Driving

Tuesday February 21 2012, 10.15am, Haldasalen, Halmstad University

Click here for a video recording of the lecture.

The concept of automated driving is not new, even though it is still regarded as futuristic.  Historical examples going back to 1939 are shown to indicate how long the idea has been considered.  The distinction between "automated" and "autonomous" is explained in order to clarify the operational concepts of driving automation, and the advantages of cooperative over autonomous automation are shown.  The distinctions between partial and full automation and between operations in dedicated lanes and in mixed traffic are also explained, in the context of practical considerations of human factors and technical feasibility.  Progress is shown on some of the important steps that have already been taken toward automated driving in PATH's experiments on cooperative adaptive cruise control, automated precision docking of transit buses and automated platooning of heavy trucks.  The most important remaining technical and institutional challenges are then identified.

About Dr. Steven Shladover

Dr. Steven Shladover is the Program Manager, Mobility at the California PATH Program of the Institute of Transportation Studies of the University of California at Berkeley.  He joined the PATH Program in 1989, after eleven years at Systems Control, Inc. and Systems Control Technology, Inc., where he led the company’s efforts in transportation systems engineering and computer-aided control engineering software products.  Dr. Shladover received all of his degrees in mechanical engineering, with a specialization in dynamic systems and control, from M.I.T., where he began conducting research on vehicle automation in 1973.  He has been active in ASME (former Chairman of the Dynamic Systems and Control Division), SAE (ITS Division) and the Transportation Research Board (Chairman of the Committee on Intelligent Transportation Systems from 2004-2010, and member of the Committee on Vehicle-Highway Automation from its founding until 2010), and was the chairman of the Advanced Vehicle Control and Safety Systems Committee of the Intelligent Transportation Society of America from its founding in 1991 until 1997.  Dr. Shladover leads the U.S. delegation to ISO/TC204/WG14, which is developing international standards for “vehicle-roadway warning and control systems”. 

Edward A. Lee on Heterogeneous Actor Models

Friday February 10th 2012, 13:00 (1pm), Wigforssalen, Halmstad University

Click here for a video recording of the lecture.

Complex systems demand diversity in the modeling mechanisms. We see this very clearly with cyber-physical systems (CPS), which combine computing and networking with physical dynamics, and hence require model combinations that integrate dynamics described using differential equations with models of software. We also see it in applications where timed interactions with components are combined with conventional algorithmic computations, such as in networked computer games. We even see it in traditional software systems when we have concurrent interactions between algorithmic components.

One way to deal with a diversity of requirements is to create very flexible modeling frameworks that can be adapted to cover the field of interest. The downside of this approach is a weakening of the semantics of the modeling frameworks that compromises interoperability, understandability, and analyzability of the models. An alternative approach is to embrace heterogeneity and to provide mechanisms for a diversity of models to interact.

In this talk, I will describe an approach that achieves such interaction between diverse models using a concept that we call "abstract semantics.” An abstract semantics is a deliberately incomplete semantics that cannot by itself define a useful modeling framework. It instead focuses on the interactions between diverse models, reducing the nature of those interactions to a minimum that achieves a well-defined composition. I will illustrate how such an abstract semantics can handle many heterogeneous models that are built today (such as Statecharts, which combine state machines with synchronous concurrent models, hybrid systems, which combine state machines with differential equations, process networks, which combine imperative programs with message passing concurrency, etc.). I will also show how it handles combinations that are not readily available in modeling tools today. I will illustrate these combinations with examples prototyped in Ptolemy II.

About Professor Edward A. Lee

Edward A. Lee is the Robert S. Pepper Distinguished Professor in the Electrical Engineering and Computer Sciences (EECS) department at U.C. Berkeley. His research interests center on design, modeling, and analysis of embedded, real-time computational systems. He is a director of Chess, the Berkeley Center for Hybrid and Embedded Software Systems, and is the director of the Berkeley Ptolemy project. From 2005-2008, he served as chair of the EE Division and then chair of the EECS Department at UC Berkeley. He is co-author of nine books (counting second and third editions) and numerous papers. He has led the development of several influential open-source software packages, notably Ptolemy and its various spinoffs. He received the B.S. degree in Computer Science from Yale University, New Haven, CT, in 1979, the S.M. degree in EECS from the Massachusetts Institute of Technology (MIT), Cambridge, in 1981, and the Ph.D. degree in EECS from the University of California Berkeley, Berkeley, in 1986. From 1979 to 1982 he was a member of technical staff at Bell Telephone Laboratories in Holmdel, New Jersey, in the Advanced Data Communications Laboratory. He is a co-founder of BDTI, Inc., where he is currently a Senior Technical Advisor, and has consulted for a number of other companies. He is a Fellow of the IEEE, was an NSF Presidential Young Investigator, and won the 1997 Frederick Emmons Terman Award for Engineering Education.