talks AND BIOS

Here is a description of the presentations and the bios of the speakers listed in the order of appearance:

1. A Project Based Approach to Kinematic Synthesis of Mechanisms

Speakers: J. Michael McCarthy, Professor, Mechanical and Aerospace Engineering, Henry Samueli Chair and Director of the Center for Engineering Science in Design, University of California, Irvine, CA

This presentation describes recent work on teaching the principles of kinematic synthesis of mechanisms through a series of projects that lead to increasingly complex leg mechanisms for walking machines. Topics include graphical constructions for the analysis and simulation of four-bar linkages, curvature theory, six-bar linkages, pantographs, and two and three position synthesis. The result is constructions for function generators that guide the joints of an RR leg to obtain a generalization of the eight-bar leg mechanism of Theo Jensen’s Strandbeest.

Biography: J. Michael McCarthy is a Professor in the Department of Mechanical and Aerospace Engineering, University of California, Irvine. He has over 200 publications in kinematic synthesis, and is the author of several books including Introduction to Theoretical Kinematics (MIT Press 1990), Geometric Design of Linkages (Springer 2000, 2011) and a new book Kinematics Synthesis of Mechanisms: a project based approach (MDA Press, 2019). Demonstrations of the constructions in this book are available on mechanicaldesign101.com.

2. Origami-based Engineering Design

Speaker: Larry Howell, Associate Academic Vice President and Professor of Mechanical Engineering, Brigham Young University, Provo, Utah

For centuries origami artists have invested immeasurable effort developing origami models under extreme self-imposed constraints (e.g. only paper, no cutting or gluing, one regular-shaped sheet). This has resulted in stunning origami structures and mechanisms that have not previously been conceived using traditional engineering methods. Using origami-inspired methods, it is possible to design origami-like systems, but using different materials and processes to meet emerging product requirements. This workshop will overview fundamentals and provide some hands-on activities to illustrate important concepts.

Biography: Larry L Howell is a Professor and an Associate Dean at Brigham Young University (BYU). Prof. Howell received his Ph.D. from Purdue University and prior to joining BYU in 1994 he was a finite element analysis consultant for Engineering Methods, Inc., and an engineer on the design of the YF-22 (the prototype for the U.S. Air Force F-22 Raptor). He is a Fellow of ASME and the recipient of the ASME Machine Design Award, ASME Mechanisms & Robotics Award, Theodore von Kármán Fellowship, NSF Career Award, among others. Prof. Howell’s research focuses on compliant mechanisms, including origami-inspired mechanisms, space mechanisms, microelectromechanical systems, and medical devices. He is the co-editor of the Handbook of Compliant Mechanisms and the author of Compliant Mechanisms. His lab’s work has also been reported in popular venues such as Newsweek, Scientific American, The Economist, Smithsonian Magazine, and the PBS documentary program NOVA.

Pre-requisites: It would be helpful if participants knew some of terms that will be used, including compliant mechanisms and some examples of origami-based engineering. This could be done by watching the following two videos:

“Why Machines That Bend Are Better,” https://www.youtube.com/watch?v=97t7Xj_iBv0

“How Origami is Inspiring Scientific Creativity,” https://www.youtube.com/watch?v=fYf7nReaGPw

3. Modeling of Linkages with Screws and Lie groups

Speaker: Andreas Mueller, Professor, Institute of Robotics, Johannes Kepler University, Linz, Austria

This lecture presents the modern approach to the kinematic modeling of linkages. It introduces the exponential mapping as the central element that enables to describe and parameterize finite motions of linkages in a compact and ‘user-friendly’ form. The first- and second-order kinematic relations are derived and applied to the velocity and acceleration analysis of open and closed loop linkages, in particular serial and parallel manipulators. To this end, the concept of frame transformation of twists and general screw coordinates is presented. The consequence of particular coordinate representations of screws and twists is discussed. A recursive O(n) formulation for computing the velocity and acceleration of a serial kinematic chain is presented. The numerical effort is discussed for the body-fixed and spatial representation. To demonstrate the actual use of the presented formulations, various examples are shown during the course. A simple Mathematica package will be available in order to allow participants to follow.

Biography: Andreas Mueller is full professor and head of the Institute of Robotic at the Johannes Kepler University Linz, Austria. Prior appointments include positions as researcher University Duisburg-Essen, Germany and at the Institute of Mechatronics, Chemnitz, Germany (also deputy CEO) and as associate professor at the Michigan University – Jiao Tong University Joint Institute in Shanghai. His research interests cover the holistic modeling and model-based control of mechatronic and robotic systems, kinematics and singularities, dynamics modeling and numerical method, mobile platforms, redundant serial and parallel kinematics manipulators, flexible lightweight robots, human machine interaction and safety.

Pre-requisites: Basic knowledge of linear algebra, vector algebra and spatial kinematics is required. Understanding of basic principles of kinematic modeling with screws is beneficial (no knowledge of ‘real’ screw theory is required).

4. Robotics and Wearables to Restore and Retrain Human Movements

Speaker: Sunil K. Agrawal, Ph.D., Professor, Department of Mechanical Engineering and Department of Rehabilitation and Regenerative Medicine, Columbia University, New York, NY 10027, USA.

Abstract: Neural disorders limit the ability of humans to perform activities of daily living. Robotics can be used to probe the human neuromuscular system and create new pathways to relearn, restore, and improve functional movements. Dr. Agrawal’s group at Columbia University Robotics and Rehabilitation (ROAR) Laboratory has designed innovative robots for this purpose using principles of parallel-actuated robots and cable-driven robotic systems. These robots have been tested on human subjects in a variety of studies to understand the neuro-muscular response. Human experiments have targeted patients with stroke, cerebral palsy, Parkinson’s disease, ALS, Vestibular disorders, elderly subjects and others. The talk will provide an overview of some of the robot designs and scientific studies performed with them.

Biography: Sunil K. Agrawal received a Ph.D. degree in Mechanical Engineering from Stanford University in 1990. He is currently a Professor and Director of Robotics and Rehabilitation (ROAR) Laboratory at Columbia University, located both in engineering and medical campuses of Columbia University. Dr. Agrawal has published more than 500 journal and conference papers, three books, and 14 U.S. patents. He is a Fellow of the ASME and AIMBE. His honors include a NSF Presidential Faculty Fellowship from the White House in 1994, a Bessel Prize from Germany in 2003, and a Humboldt US Senior Scientist Award in 2007. He is a recipient of 2016 Machine Design Award from ASME for “seminal contributions to design of robotic exoskeletons for gait training of stroke patients” and 2016 Mechanisms and Robotics Award from the ASME for “cumulative contributions and being an international leading figure in mechanical design and robotics”. He is a recipient of several Best Paper awards in ASME and IEEE sponsored robotics conferences. He has held positions of a Distinguished Visiting Professor at Hanyang University in Korea, a Professor of Robotics at the University of Ulster in Northern Ireland, a Visiting Professor at the Biorobotics Institute of SSSA in Pisa, and a Visiting faculty at Peking University. He has successfully directed 26 PhD student theses and currently supervises the research of 10 PhD students at ROAR laboratory at Columbia University in New York City.

5. Quaternions, Dual Quaternions, and Clifford Algebra

Speaker: Dr. Jeff Ge, Chair and Professor, Mechanical Engineering, Stony Brook University, NY

Abstract: This lecture introduces the fundamentals of quaternions, dual quaternions, double quaternions, Clifford Algebra, and explores their applications in computer aided design of freeform rational motions as well as data-driven simultaneous type and dimensional synthesis of freeform algebraic motions as applied to linkages and parallel manipulators.

Biography: Jeff Ge's current research areas include Computer Aided Design and Manufacturing (CAD/CAM), Robotics and Manufacturing Automation, Mechanical System Design and Simulation, and Application of Geometric Modeling and Computer Graphics in Engineering. He is one of the first to bring together the fields of Geometric Modeling and Kinematics to develop a computational-geometric framework for motion interpolation and approximation. He was Program Chair (2007) and Conference Chair (2008) ASME Mechanisms and Robotics Conference. He served as General Technical Program Chair of 2006 and 2008 ASME International Design Engineering Technical Conferences (IDETC) and Computers in Engineering (CIE) Conferences. He was an Associate Editor of ASME Journal of Mechanical Design (2003-2006) and ASME Journal of Mechanisms and Robotics (2010-2016). He had been on the Editorial Boards of International Journal of Mechanics Based Design of Structures and Machines, International Edition of Chinese Journal of Mechanical Engineering, as well as ASME Book Series on Robotics Engineering. He is a Fellow of ASME, was Chair of ASME Mechanisms and Robotics Committee, and currently Chair of ASME Design Engineering Division. Jeff Ge was also Chair of US Council on the Theory of Mechanisms and Machine Theory, Chief US Delegate to 2015 IFToMM World Congress and has been elected Chair of the Constitution Committee of IFToMM in 2019.

6. Type Synthesis and Reconfiguration Analysis of Multi-mode Parallel Mechanisms

Speaker: Dr. Xianwen Kong, Heriot-Watt University, UK

Abstract: Multi-mode parallel robots, which can act as two or more conventional parallel robots, are one class of reconfigurable robots that have attracted attention in the past two decades. This lecture starts with the type synthesis of multi-mode parallel mechanisms and then focuses on the reconfiguration analysis of multi-mode mechanisms. Reconfiguration analysis of a mechanism is to identify all the operation modes of the mechanism and transition configurations among different operation modes. To facilitate understanding of motion characteristics of a multi-mode parallel mechanism under different operation modes, the kinematic interpretation of quaternions with different number of zero components is presented. Examples are given to illustrate how to obtain operation modes and transition configuration of multi-mode parallel mechanisms by solving kinematic constraint equations using tools from computer algebraic geometry (such as SINGULAR and Maple).

Biography: Dr. Xianwen Kong is a lecturer at Heriot-Watt University, UK. His research focuses on the creative design of parallel manipulators with their applications in manufacturing and renewable energy. He has authored or co-authored one monograph (with Prof C. Gosselin), two US patents and more than 200 publications in journals and conference proceedings. The Russian translation and Chinese translation of the monograph were published in 2012 and 2013 respectively. Dr Kong served as the Program Chair/Co-chair for the ASME/IEEE ReMAR in 2009, 2012 and 2015 and the ASME IDETC 2016 and 2018. He is an ASME fellow and a member of the ASME Mechanisms and Robotics Committee and serves as an associate editor for ASME J Mech Rob and Mech Mach Theory, a member of editorial board of the Chinese J Mech Eng. He received several awards including the 2012 ASME Freudenstein/General Motors Young Investigator Award.

Pre-requisites: Basic knowledge spatial kinematics is required. SINGULAR (https://www.singular.uni-kl.de/) is recommended. You could install it on your laptop or access the software online.

A slide with the animation of a multi-mode parallel mechanism can be downloaded from

http://home.eps.hw.ac.uk/~xk5/KiSS2019/ .

Recommended reading (optional): https://www.sciencedirect.com/science/article/pii/S0094114X1300253X

7. Machine Learning Fundamentals for Kinematics and Robotics

Speaker: Nilanjan Chakraborty, PhD, Assistant Professor, Mechanical Engineering at Stony Brook University, Stony Brook, NY

Abstract: This lecture will give a quick overview of current deep learning techniques. The lecture will be self-contained with a brief introduction to supervised machine learning. Topics to be covered include linear regression, neural networks, convolution neural networks (CNN), variational auto-encoders and Generative Adversarial Networks. Applications of these techniques in the context of mechanisms and robotics will also be discussed. Prerequisites include familiarity with basic linear algebra and unconstrained optimization.

Biography: Nilanjan Chakraborty is currently an Assistant Professor in Mechanical Engineering at Stony Brook University, Stony Brook, NY. Before joining Stony Brook, he was a Post-doctoral researcher and project scientist at the Robotics Institute, Carnegie Mellon University, Pittsburgh, PA. He obtained his PhD in Computer Science from Rensselaer Polytechnic Institute, Troy, NY. He has a MSc (Engg.) degree in Mechanical Engineering from Indian Institute of Sciences, Bangalore, and a BE degree in Mechanical Engineering from North Bengal University. His research interests are in robotics, artificial intelligence, dynamical systems, and applied optimization. He is currently an Associate Editor for IEEE Robotics and Automation Letters. He was a recipient of the Best Paper Award (Computer Systems Technical Group) at the Annual Meeting of the Human Factors and Ergonomics Society in 2013 and the Best Student Paper Award at Robotics: Science and Systems (RSS) Conference in 2007.

Prerequisites: include familiarity with basic linear algebra and unconstrained optimization.

8. Machine Learning for Kinematic Synthesis of Mechanisms

Speaker: Anurag Purwar, PhD, Research Associate Professor, Mechanical Engineering, Stony Brook University, Stony Brook, NY

Abstract: Decades of research in mechanism synthesis has led to a body of literature rich in algorithms and computational techniques for kinematic synthesis of mechanisms and robots. It is said that true knowledge resides in the questions that we ask. In that regard, mechanism synthesis work has largely depended on framing questions in a way that contrives to seek answers, which can be obtained mathematically or computationally. This is visible starkly in specification of the input to the synthesis problems, often posed as precision point or position problem, which makes it difficult to obtain good, defect-free solutions to the problems. In this presentation, we will present deep generative learning models, such as Variational Auto-Encoder (VAE) and Conditional Variational Auto-Encoder (C-VAE) to solve problems that have had no good theoretical underpinning, such as defect-free generation, conditioning of the input, and contextual concept generation of mechanisms.

Time permitting, attendees will also learn how to design and prototype robot motions using a novel motion design app at http://www.motiongen.io and a robot design kit called SnappyXO (www.snappyxo.com). This would be a completely hands-on session where in attendees will prototype a linkage-based robot. This part of the presentation aligns with the first talk of the summer school by Prof. Mike McCarthy.

Biography: Dr. Anurag Purwar is an award-winning teacher, researcher, TEDx speaker, and inventor of several technologies, some of which are available as products in the market. He has received several best paper and outstanding research awards, excellence in teaching awards, and the top 100 design awards for his inventions. He received the SUNY FACT2 award, two SUNY Research Foundation Technology Accelerator Fund (TAF) awards, A.T. Yang award for Theoretical Kinematics, and Presidential Award for Excellence in Teaching.

Dr. Purwar has led more than 125 technical projects in mechanisms and robotics, wave-, and wind-energy harvesting, physical therapy and rehabilitation devices, aircraft components, and consumer products design, with a cumulative in-cash funding of $4.4M supported by National Science Foundation, industry, NY-state SPIR, NY-state Center for Biotechnology, Sensor-CAT, SUNY Research Foundation, and SUNY Office of Provost. More than 175 students have been supported on these projects.

Dr. Purwar is a Research Associate Professor of Mechanical Engineering at Stony Brook University. He is currently an Associate Editor of the American Society of Mechanical Engineer (ASME) Journal of Computing and Information Science in Engineering and of International Journal of Mechanics Based Design of Structures and Machines and has served as the Conference and Program chair for several ASME international conferences. He is an elected member of the ASME Mechanisms and Robotics Committee and of the National Academy of Inventors (NAI).

Prerequisites: Laptop with internet, familiarity with basic linear algebra and optimization.

9. Optimization-Based Synthesis – Shrinking the Search Space to Find Better Mechanisms

Speaker: Jim Schmiedeler, Professor of Aerospace and Mechanical Engineering, University of Notre Dame, Notre Dame

Abstract: Mechanism designers continue to make extensive use of the ever-improving computational power available to them. The reality, though, is that the computational power will never be enough. The complexity of the problems mechanism designers seek to solve advances in lock step with the gains in computational resources. When exhaustive searches of the design space are either infeasible in the allotted time (if at all) or unjustified by the problem (too costly for the benefit to be realized), optimization-based approaches are preferred. Of course, this trades the computational challenge of complete design space searches for the difficulties of avoiding local minima. This presentation will focus on strategies that have proven useful in optimization-based synthesis of both rigid-body and compliant mechanisms. The general approach is to leverage as much of the designer’s a priori knowledge of the problem as possible to intelligently narrow the search space, effectively bootstrapping the optimizer such that superior solutions can be found in less time.

Biography: Jim Schmiedeler is a Professor and Associate Department Chair in Aerospace and Mechanical Engineering at the University of Notre Dame. He received the B.S. degree from Notre Dame and the M.S. and Ph.D. degrees from The Ohio State University, all in mechanical engineering. He was previously an Assistant Professor at the University of Iowa and at Ohio State and a summer faculty research fellow at NASA’s Jet Propulsion Laboratory. He is a recipient of the Presidential Early Career Award for Scientists and Engineers (PECASE) and a fellow of the American Society of Mechanical Engineers.

Prerequisites: none

Report abuse