Scientific Symposium
Location: MPI-IS Stuttgart room 2R04
All current and former employees and partners of the Max Planck Institute for Intelligent Systems are welcome to attend this event.
If you have any questions, please contact Eva Lämmerhirt, Institute Management Officer, at eva.laemmerhirt@tuebingen.mpg.de
Schedule
Wednesday, January 25, 2023 |
||
---|---|---|
Time | Activity | |
8:30-9:00 |
Welcome by Prof. Metin SittiManaging Director |
|
09:00 | Talks (25 min. talk, 5 min. Q&A) |
|
9:00-9:30 |
Martin SchmitzSaarland University Interweaving Atoms and Bits: Using 3D Printing to Create Tailored Human-Computer Interfaces Chair: Katherine J. Kuchenbecker Abstract and speaker’s biography >> AbstractWith increasing digitization, computer science is steadily making its way into intelligent products, as well as their increasingly autonomous production, logistics, and use. Examples range from simple smart devices for the home to full-fledged manufacturing robots, which both increase in number every day. For the successful integration of the rising number of intelligent products into a sustainable society, however, the quality of interaction between humans and computers is one decisive key. Simultaneously, production gets increasingly digitized via 3D printing. By enabling broad masses of users and companies to invent previously impossible or highly customized products, precisely tailored to the custom needs of users or use cases, 3D printing is considered to have the potential to trigger a new industrial revolution. A recent stream of research aims to leverage the power of 3D printing to better interweave the rising number of intelligent products into everyday life by creating precisely tailored human-computer interfaces. While advances in materials science and engineering have brought us closer to this vision, a myriad of research challenges remains in the area of human-computer interaction, for example, how to design, manufacture and use such 3D-printed intelligent products, or how to find suitable 3D-printed interactive structures and interaction concepts. The talk will present past and current contributions to this vision. BiographyMartin Schmitz is a senior lecturer and researcher at the human-computer interaction lab of Jürgen Steimle at Saarland University. Previously, he was a postdoctoral researcher and group leader at the telecooperation lab of Max Mühlhäuser at TU Darmstadt, where he also received his doctoral degree. Additionally, he has been an affiliated researcher at the University of Glasgow and a visiting researcher at the University of Copenhagen. His work is regularly published and awarded at CHI and UIST, two of the most prestigious international venues in the field of human-computer interaction. His research interests are at the intersection of human-computer interaction and digital fabrication, but also include extended reality and haptics. In general, he enjoys designing, building, and evaluating human-computer interfaces that strive to be tailored to individual users or use cases rather than to a mass-produced common denominator. |
|
9:30-10:00 |
Małgorzata Włodarczyk-BiegunSilesian University of Technology, in Gliwice, Poland, and The University of Groningen, The Netherlands Melt electrowriting of hierarchical structures to reconstruct native tissues Chair: Katherine J. Kuchenbecker Abstract and speaker’s biography >> AbstractNative tissues are highly organized structures, with an intricate cellular microenvironment. They are non-homogeneous and typically exhibit functional gradients in the architecture and biochemical composition. Whereas it is challenging to produce biomimetic scaffolds using traditional manufacturing processes, 3D printing has emerged as a perfect tool to construct complex 3-dimensional objects with great precision. In a variety of available printing approaches, Melt Electrowriting (MEW) is especially interesting due to the high resolution, effectiveness, and low costs of the process. In this lecture, I will present our work on 3D printing of hierarchical structures, obtained with high precision at very small scales using MEW. MEW combines principles of electrospinning with 3D printing, allowing precise deposition of a molten polymer as fibers with diameters in the micrometer range, with high flexibility of the design. By alternating scaffolds’ architecture, the material systems with the structural, mechanical, and biological properties relevant to different tissues can be obtained. Particularly, I will present the utility of MEW for the reconstruction of the Human Trabecular Meshwork (membrane located in the eye), hard-soft tissue interfaces (e.g. connection between bone and tendon), and skin tissue, with a focus on the scaffold design. The proposed small-scale biomimetic scaffolds hold great promise for future high-accuracy in-vitro testing models or low-cost patient-specific implantable systems. BiographyMałgorzata Włodarczyk-Biegun is an Assistant Professor at the Silesian University of Technology, in Gliwice, Poland, and a Principal Investigator at The University of Groningen, The Netherlands. Her research is focused on applying 3D (bio)printing techniques to generate complex hierarchical scaffolds of polymeric materials for advanced tissue regeneration. She is also interested in the rational design and development of novel polymer-based bioinks with tunable properties allowing to induce a specific cellular response, to produce systems with bio-instructive properties. She has earned her Master's degree in Poland, in Psychology, and in Biomedical Engineering. For her PhD, she decided to continue in the field of Biomedical Engineering. She obtained the title in January 2016, at Wageningen University and Research Center in the Netherlands, working on recombinant proteins for biomedical applications. In the years 2016-2020, she was employed at INM-Leibniz Institute for New Materials, in Germany as a postdoctoral researcher. There, she took on the challenge to set up from scratch and supervise the Bioprinting Lab. In 2017, she was awarded a UNESCO-L’Oréal for Women in Science Prize in recognition of her research qualities and ability to combine excellence in Science with being a mother. Currently, she is working in the Polymer Science group at the University of Groningen, where she has received 3 prestigious grants from NWO (Netherlands Organization for Scientific Research). She has also been awarded NAWA Polish returns grant (2019) intended to set up her own research group in Poland, and the polish NCN Opus grant to further support her research. Combining all sources of funding, she is now leading a young Research Group, located partially in the Netherlands and partially in Poland, to pursue international collaborative work. |
|
10:00-10:30 |
Kento YamagishiNorthwestern University Tissue-adhesive ultrathin electronics for bio-conformable wearable and implantable systems Chair: Katherine J. Kuchenbecker Abstract and speaker’s biography >> AbstractDespite the rapid research progress in flexible and stretchable electronic materials and devices, it remains challenging to realize the comfortable and long-term use of bio-integrated wearable and implantable devices interfaced with the animal/human body. Most of the existing implantable devices require invasive procedures such as suturing to secure the device onto the tissues. Such invasive procedures may affect the original motility and function of the organs, and in a worse case, cause serious inflammation. To address this issue, I have investigated the use of polymeric and elastomeric ultrathin films with thicknesses less than 10 µm as a highly conformable and adhesive bio-interfacing platform to achieve suture-less device fixation and mitigate the mechanical mismatch between rigid electronics and soft biological tissues1. In this presentation, I discuss the development of tissue-adhesive ultrathin-film electronics and their applications as wearable and implantable systems, highlighting the following three demonstrations; (1) conductive nanosheet-based skin-contact electromyography devices for unexplored muscles in athletes2, (2) tissue-adhesive wireless optoelectronic devices for fully implantable cancer therapeutics3,4, and (3) ultra-conformable strain sensors for monitoring gastrointestinal motility. The tissue-adhesive ultrathin electronics will form the novel foundations of technology for building human-machine interactive intelligent systems. BiographyKento Yamagishi, Ph.D. is a postdoctoral fellow at the Center for Bio-Integrated Electronics at Northwestern University, USA. He received his Ph.D. in Engineering from Waseda University, Japan, in 2018, where he held the Research Fellowship of Japan Society for the Promotion of Science. He was a postdoctoral scholar at the Research Organization for Nano & Life Innovation at Waseda University (2018) and at the Digital Manufacturing and Design (DManD) Centre at Singapore University of Technology and Design (2018-2021), and then moved to Northwestern University (2021-). His research is focused on soft materials, biomaterials, flexible and stretchable electronics, and digital fabrication for the development of wearable and implantable biomedical devices. His honors include Abe Research Award for Young Scientists from Japanese Society for Medical Engineering (2018), the 2nd prize for the International Federation for Medical and Biological Engineering (IFBME) Young Investigator Award (2019), and nominee for Best Poster Award in 2019 MRS Fall Meeting & Exhibit (Boston, 2019) and 2022 MRS Spring Meeting & Exhibit (Honolulu, 2022). |
|
10:30-11:00 | Break |
|
11:00-11:30 |
Pia BideauTechnical University Berlin Learning for and from motion Chair: Michael Black Abstract and speaker’s biography >> AbstractPerceiving motion is a fundamental ability that humans learn similar to walking or talking. Also, humans need to learn how to process and to use perceived motion information in order to understand their environment and to interact with it appropriately. The perception and processing of motion reveals interesting object properties like for example “rigidity”, “distance”, “speed”, or even higher-level “actions” like running or biking - to name just a few. While humans naturally learn from interacting with the environment and other peers, recent computer vision research surprisingly has not focused on such interaction-informed perception. We therefore ask: How does action matter for vision? More specifically, how does action shape learned visual representations? My work aims at discovering and making use of information that arises at the intersection of vision and action. This appears to be an incredibly rich source of information, ranging from physical relations between motion and image formation to learned high level visual representations shaped by action understanding. I believe such information lying at the edge of computer vision and robotics, will not only help to develop more robust artificial vision systems, furthermore it will allow learning with only minimal human supervision. In this talk I will lay out my earlier and current research, which may be categorized along these broad themes: (1) achieving object understanding from motion, and (2) learning to generate motion (e.g. trajectories) by inducing priors about high-level action understanding. BiographyPia Bideau is postdoctoral researcher at Technical University Berlin and the Excellence Cluster Science of Intelligence. Before joining Science of Intelligence in January 2020, Pia Bideau was a PhD student at the University of Massachusetts, Amherst. Her thesis proposed novel approaches towards segmenting independently moving objects from noisy optical flow estimates. She studied at Ruhr-University Bochum where she received her Master in Electrical Engineering and Information Technology. Pia Bideau has been co-organizing the workshop “What is motion for?” at ECCV 2022. Currently she is postdoctoral representative at the Excellence Cluster Science of Intelligence. She received a best paper award at the ECCV 2018 workshop "What is optical flow for?" for her paper “MoA-Net: Self-Supervised Motion Segmentation” and scholarships from DAAD for academic education abroad and BMBF for her excellent academic achievements during her Master studies. |
|
11:30-12:00 |
Yanpei HuangImperial College London Movement augmentation for robot-assisted surgery Chair: Michael Black Abstract and speaker’s biography >> AbstractMany surgical tasks require using three or more tools simultaneously, which is why the DaVinci Teleoperated Surgical System has four robotic arms. Currently, surgeries are typically performed by a main surgeon and an assistant, where it is known that their performance can be affected by miscommunications. Providing the surgeon with tools and techniques to control three or four surgical tools by themself would avoid miscommunication and could improve the outcome of surgery. However, the current interfaces allowing a surgeon to control several tools offer only limited precision, dexterity and intuitiveness. Therefore, I developed a dedicated foot interface and techniques that allow an operator to control three tools intuitively and continuously. Using individual motion patterns, the operator could simultaneously control a novel flexible endoscope and its two tools, as was tested in an intervention on a pig stomach. In addition, I used my interface to systematically investigate the main characteristics, opportunities and limitations of movement augmentation for "three-handed manipulation". In my presentation, I will present these works and outline the opportunities that movement augmentation opens up for applications in industry and medicine, especially for solo surgery. BiographyDr. Yanpei Huang's main research interests are in the development of intuitive human-machine interaction and in medical robotics. Yanpei earned a BSc in mechanical design, manufacturing and automation from Southwest University, China, in 2010, with a second BSc degree in Law. Subsequently, she worked as a design engineer from 2010 to 2011 and from 2012 to 2014 as an attorney. She then started graduate studies at the Nanyang Technological University (NTU) in Singapore, where she obtained an MSc in manufacturing systems and engineering in 2016. During her Ph.D. at NTU, she developed interfaces and techniques to enable a surgeon controlling a soft endoscope with two tools without requiring an assistant. In 2020, she joined Imperial College London as a post-doctoral Research Associate in the Human Robotics Group, where she investigates movement augmentation strategies in Virtual Reality. |
|
12:00-12:30 |
Alexandra MoringenBielefeld University Modeling of Learning and Interaction for Humans and Robots Chair: Michael Black Abstract and speaker’s biography >> AbstractIn this presentation I will discuss research opportunities that are created by bringing together sensor technology, study of cognitive interaction, and artificial intelligence. The presented projects address research questions in three domains: 1) How to disentangle continuous multimodal human interaction time series into primitive actions, to enable us to better evaluate and analyse e.g. learning progress, or to give the most suitable feedback cues? 2) How to enable robot learning of skills with multimodal perceptual modalities, such as haptics? How to enable a robot to employ optimal primitive actions and efficiently process multimodal cues during this process? 3) And finally, how to build and optimize computational teaching systems that accelerate human learning of skills for which guidance by experts is otherwise necessary? How to split learning of complex skills into primitive practice entities, and choose the ones with the highest degree of usefulness for a given individual? To this end, I present multiple novel experimental frameworks and corresponding findings. BiographyAlexandra Moringen studied mathematics and computer science at the Christian-Albrecht University of Kiel. She obtained her PhD in Intelligent Systems from the Bielefeld University. There she co-leads a bilateral project financed by the German Israeli Foundation for Scientific Research and Development. Her work focuses on meta-learning and acceleration of learning with computational scaffolding. |
|
12:30-13:30 | Break |
|
13:30-14:00 |
Shuo LiNorthwestern University Design and Manufacturing of Functional Soft Materials for Bioinspired Robotics and Bio-integrated Electronics Chair: Metin Sitti Abstract and speaker’s biography >> AbstractFunctional soft materials are of great and increasing interest for applications ranging from soft robotics to biomedical devices. Advancing machine intelligence by way of engineering new materials yields mechanical and physicochemical properties towards unprecedented robotic capabilities and bioelectronic functionalities. In this talk, I will first introduce the design, synthesis, and reliable manufacturing of robotic materials for actuation, perception, signaling, and their integration. Several examples will be highlighted, including liquid crystal elastomer artificial muscles, stretchable optoelectronic sensors, and hyperelastic light-emitting displays. I will also describe a new approach that actively combines bioresponsive hydrogels with implantable electronic systems for continuous, wireless, physiological biosensing towards translational biomedical research and unmet clinical needs. BiographyShuo Li received his B.S. degree from the Univeristy of Illinois Urbana-Champaign, and the M.S. and Ph.D. degrees from Cornell Univeristy, all in Materials Science and Engineering. He is currently a Postdoctoral Fellow at the Querrey Simpson Institute for Bioelectronics at Northwestern University. His research focuses on using stimuli-responsive soft matter, bioinspired design approaches, in combination with advanced manufacturing methods for unusual, material-driven device systems that create new functional opportunities in the fields of robotics and bio-integrated electronics. He is a recipient of the Extraordinary Potential Prize of 2019 Chinese Government Award for Outstanding Self-financed Students Abroad. His work has been featured in popular media outlets such as the Washington Post, Scientific American, and NBC News. |
|
14:00-14:30 |
Yoav MatiaSibley School of Mechanical and Aerospace Engineering, Cornell University From Robotic Towards Autonomous Materials Chair: Metin Sitti Abstract and speaker’s biography >> AbstractOver the last twenty years, progress in soft robotics has motivated the materialization of physical intelligence through new forms of soft actuators, sensors, and control strategies. Nevertheless, the challenges of performance, and control in soft robotics remain limited by available materials. The natural world is inspiring a new generation of materials design paradigms, where multifunctionality not only tightly integrates multiple robotic capabilities, but enables truly autonomous behaviors. In this talk, I present my work on morphological intelligence, the embedding of logic in otherwise inert structural material, to organically merge control functions into the material without adding technological complexity. Next, I present an investigation of a novel approach to making a system with internal moving parts continuously function under deformation without jamming; showing its utility in a scalable elastomeric pump designed specifically to satisfy the needs of soft robotics. Last, I present these developments as fundamental precursor mechanisms for emergent material-based autonomy and agency where materials act on their own behalf. Part of a future vision of autonomous materials where information, agency, and computation are embedded into the material by virtue of internal architecture and physical mechanisms rising from multiphase multiphysics interactions organic to both the systems and the manufacturing method – The next generation of autonomous agents should be indistinguishable from materials, and integrate distributed actuation, perception, control, and energy capabilities. BiographyYoav Matia has been a postdoctoral research associate at Cornell University since 2019 and was awarded the distinguished Zuckerman STEM leadership Research fellow, and the US Army Research Laboratory, Research Associate Program Fellow. He received a Bs.c. in Mechanical Engineering from Afeka 2010 (Cum Laude), and his M.sc. and Ph.D. in Mechanical Engineering from the Technion 2019, with honors of the Pnueli Award, in recognition of the best doctoral research conducted at the Faculty of Mechanical Engineering, Technion-IIT. Prior to his Ph.D., he worked in Israel’s High-Tech industry from 2001-2014, the last eight of which were as an R&D engineer. His research covers the interdisciplinary field of fluid-structure interactions, where the interplay between fluid dynamics and solid mechanics in a host of non-linear transient phenomena governs the physics of the system. In his work, Yoav aims to balance theoretical modeling, experimental illustrations, and numerical simulations to promote and enlighten design and technologies. |
|
14:30-15:00 |
Philipp RothemundMax Planck Institute for Intelligent Systems Intelligent Soft Systems for the Robots of the Future Chair: Metin Sitti Abstract and speaker’s biography >> AbstractFunctional soft materials promise to revolutionize robotics hardware by providing new functions and by allowing robots to perform better than traditional robots made from hard materials such as metals and electromagnetic motors. In particular, they allow implementing functions in entirely new ways compared to traditional machines when their nonlinear material properties and multiphysics interactions are utilized. Today, we are only at the beginning of exploiting the opportunities that are presented by these new materials. In this talk, I discuss physical effects in soft materials, which can enable high-performing soft devices. I describe how snap-through instabilities in elastomeric structures can be exploited for electronics-free control of soft robots. I also discuss how the electromechanical coupling between electric fields and soft structures enables high-performing electroactive devices. BiographyDr. Philipp Rothemund is a postdoctoral researcher in the Robotic Materials Department at the Max Planck Institute for Intelligent Systems, where he studies the fundamental physics of electrohydraulic actuation. During 2019-2020, he was a postdoctoral researcher at the University of Colorado Boulder. Before starting his postdoc, he lectured multiphysics simulations at Universidad Pontificia Comillas in Spain during 2018. Philipp earned his Ph.D. and M.S. in Engineering Sciences from Harvard University in 2018 and 2014, respectively, by studying the mechanics of soft actuators. He also received an M.S. and B.S. in Mechanical Engineering from ETH Zurich in 2012 and 2009, respectively. His research interests include soft robotics, energy harvesting, nonlinear continuum mechanics, multiphysics modeling, and materials science. |
|
15:00-15:30 | Break |
|
15:30-16:00 |
Cameron AubinCornell University Towards Enduring Autonomous Robots via Embodied Energy Chair: Christoph Keplinger Abstract and speaker’s biography >> AbstractModern robots lack the combination of endurance (the ability to operate for long durations) and adaptability (the ability to perform a multitude of life-sustaining functions across different environments) that we see in living organisms. One explanation for this phenomenon is that robots are typically composed of individual power, actuation, sensory, and control blocks, each existing as separate materials and optimized for different tasks. Conversely, biological systems are highly interconnected at the material (cellular) level, are hierarchical, and capable of self-assembly, thereby enabling increased complexity and multifunctionality. Many robots also house energy in a single storage unit, often a battery, while more advanced organisms distribute energy throughout their entire bodies. In this talk, I will discuss my work developing robots with multifunctional embodied energy systems that increase their performance, endurance, and adaptability. My discussion will focus on the creation of an untethered soft robotic fish containing a “synthetic vascular system” inspired by redox flow batteries. This system combines the functions of mechanical actuation with hydraulic force transfer and electrolyte-based energy storage, increasing the energy density of the robot by over 300%. Next, I will discuss how energy-dense chemical fuels can be used to create powerful (Pstroke > 90 W, Fstroke > 9 N, Pspec = 147 W/g), soft microactuators. We have leveraged microscale combustion and localized flame quenching to develop two demonstrations: a soft haptic display that operates at high frequencies (ƒ > 1 kHz) without electromechanical valves, and a quadrupedal microrobot capable of multigait/directional movement. Using energy-dense fuels at this scale can increase operating time, power density, and ultimately untether microrobotic systems. BiographyCameron is a PhD candidate in the Department of Mechanical and Aerospace Engineering at Cornell University. He is a member of the Organic Robotics Laboratory and is advised by Dr. Robert Shepherd. Cameron’s work centers on improving the endurance, adaptability, and autonomy of robots by through the integration of multifunctional, biologically-inspired energy systems. His interests also include soft robotics, advanced materials manufacturing, and microrobotics. Cameron earned a B.S.E in Biomedical Engineering from Duke University in 2014 and an M.S. in Mechanical Engineering from Cornell University in 2019. His is a recipient of the Harriet Davis Fellowship and his research has been featured in several popular media outlets, including CNN, PBS, BBC, Wired, and more. |
|
16:00-16:30 |
Aniket PalMax Planck Institute for Intelligent Systems Soft materials enabling intelligent systems: from robotics to biomedical applications Chair: Christoph Keplinger Abstract and speaker’s biography >> AbstractSoft materials have gained increasing prominence in science and technology over the last few decades.This shift from traditional rigid materials to soft, compliant materials have led to the emergence of anew class of devices which can interact with humans safely, as well as reduce the disparity inmechanical compliance at the interface of soft human tissue and rigid devices. In this talk, I present avariety of applications of soft materials, from wearable biosensors to soft robotics and mechanicalcomputing. Starting with biosensors, I show how inexpensive polymeric materials can be used toimprove upon as well as develop entirely new wireless sensors to measure and communicate the levelof various analytes in and the properties of multiple biofluids: blood, urine, wound exudate, and sweat.Along with sensors, another prominent area of soft materials application has been in actuators androbots which mimic biological systems not only in their action but also in their soft structure andactuation mechanisms. In this talk I show design strategies to improve upon current soft robots byintroducing intelligence in the physical morphology of soft robots. Specifically, I present two methods:programming the storage of elastic strain energy and tuning the tessellation architecture of softactuators, to create programmable deformations and motions. Finally, I present the use of mechanicalinstabilities to develop entirely mechanical, monolithic computational platforms for seamlessinteractions with the environment and binary logical computations with mechanical signals. Thecomputational systems are also made programmable by carefully introducing magnetic properties inthem and using an external magnetic field for control. Collectively, this talk demonstrates the use ofsoft compliant materials as the foundation for developing new sensors and actuators for human use andinteraction. BiographyAniket Pal is currently a Humboldt Postdoctoral Fellow at the Physical Intelligence Department, MaxPlanck Institute for Intelligent Systems (Stuttgart). He received his B.Eng. degree from the Departmentof Production Engineering, Jadavpur University in 2014. He then joined the School of IndustrialEngineering at Purdue University, where he worked on soft robotics and flexible biosensors, andobtained his PhD in 2020. His current research interests focus on modeling and development of softrobots, mechanical metamaterials, and flexible electronics. |
|
16:30-17:00 |
Florian HartmannEPFL Challenges and Opportunities in Soft Robotics: A Sustainable Approach Chair: Christoph Keplinger Abstract and speaker’s biography >> AbstractResearchers are able to create ever more life-like robots by drawing inspiration from natural designs, structures, and functionality. However, sustainability aspects have been widely neglected in this process. Establishing eco-friendly materials, efficient use of energy, and non-toxic fabrication methods for robotics and electronics is crucial to reduce growing amounts of tech-waste. Combining high performance, such as durability and good mechanical and electrical properties, with environmentally-friendly design or biodegradability presents the key challenge of this endeavor. This talk focuses on sustainable material choices for soft robotics and electronics and addresses questions such as "how can materials and design significantly boost performance" to enable devices with longer lifetimes but that still degrade after use. Performance and sustainability can both be achieved in new technologies if we make the effort to design materials and devices accordingly. BiographyFlorian Hartmann is a physicist with a Doctor in Technical Sciences from the Johannes Kepler University Linz in Austria. He specializes in soft robotics and electronics, flexible electronics, and sustainable materials, and has published extensively on these topics. During his doctoral studies Florian Hartmann has worked at Linköping University in Sweden as a visiting researcher. He is currently a postdoctoral researcher at the École Polytechnique Fédérale de Lausanne (EPFL), Switzerland, where he is working with Prof. Herbert Shea on small-scale swimming robots. |
|
17:00-17:05 | Closing remarks, Prof. Metin Sitti, Managing Director |
Organizers
Referentin der Geschäftsleitung | Institute Management Officer
Tübingen | |
+49 7071 601 1766 | |