Nix Award and Lecture Symposium: Learning from Nature – From Insight to Sustainable Innovation: Nix IV Award Lecture: From Bioinspiration to Machine Learning – A New Concept for Object Manipulation
Sponsored by: TMS Materials Processing and Manufacturing Division, TMS: Nanomechanical Materials Behavior Committee
Program Organizers: Wendelin Wright, Bucknell University; Gang Feng, Villanova University
Wednesday 8:30 AM
March 22, 2023
Room: Sapphire D
Location: Hilton
Session Chair: Wendelin Wright, Bucknell University; Gang Feng, Villanova University
8:30 AM Invited
2023 William D Nix Award Lecture: From Bioinspiration to Machine Learning – A New Concept for Object Manipulation: Eduard Arzt1; 1INM – Leibniz Institute for New Materials and Saarland University
Evolution has evolved fascinating resource-efficient and sustainable materials architectures to ensure survival. Inspired by nature, micropatterning of polymeric surfaces has become a powerful paradigm: celebrated examples range, e.g., from controlled wetting and anti-icing to coloration and switchable adhesion. Fundamental adhesion studies have not only demonstrated the benefit of microfibrillar architectures but have also inspired innovative pick-and-place systems or delicate adhesives for skin and body organs. But several problems remain: how do we release micro-objects with negligible mass? And how can we ensure reliability of gripping, also in demanding conditions such as in space? We have proposed a machine-learning based optical monitoring system that images the individual fibrillar contacts in operando. Several classifiers predict successful handling with a high accuracy, indicating, e.g., incomplete or off-center gripping. The improved reliability of this technology will impact everyday life as eco-friendly solutions will be increasingly essential for our own survival.
9:30 AM Invited
Deep Learning from Nature and Machines for Engineered and Biological Materials: Subra Suresh1; 1Nanyang Technological University
Rapid advances in computing and machine learning, along with developments in materials processing that mimic unique features and characteristics found in natural systems, offer unprecedented opportunities to design and deploy a new generation of engineered and biological materials. We demonstrate specific examples of materials design, analysis, or characterization using machine learning and biomimetics from the vantage point of three different disciplines: materials science, plant science, and biomedical science. The examples chosen here cover a spectrum of topics that include: modulating the bandgap of natural and engineered materials for applications in microelectronics, optoelectronics and energy systems; characterization of material properties at multiple length scales using multi-fidelity approaches in machine learning; design of new classes of plant-based materials with unique properties for environmental sustainability, soft robotics and flexible electronics; and machine-learning approaches combined with microfluidics and computational simulations to assess, monitor and guide clinical outcomes in such diseases as diabetic retinopathy.
10:10 AM Break
10:30 AM Invited
Bioinspired Designs for Micro-object Releasing: Xuan Zhang1; 1Leibniz Institute for New Materials
Microstructured surfaces have become an innovative and sustainable strategy for robotic gripping and handling. One crucial problem appears in micro robotic handling: the release and precise placement of superlight micro-objects which, for dimensional reasons, tend to stick tenaciously to other surfaces. As the objects shrink in size, their gravitation will be too small to allow detachment by conventional mechanisms. Here, we have proposed several bioinspired designs for controlled release mechanisms of micro-objects. Following theory, numerical simulations, and demonstration experiments, fibirils with geometry modifications arranged in regular pattern, and a metastructure involving a snap-action were developed. These innovative designs proposed, due to the purely mechanical trigger for realizing the release and unprecedented high switching ratio ~104, have the potential to create a newly energy-efficient paradigm for handling and placing superlight objects.
11:00 AM Invited
Artificial Muscles for the Lifelike Robots of the Future: Christoph Keplinger1; 1Max Planck Institute for Intelligent Systems
Robots today rely on rigid components and electric motors, making them heavy, unsafe near humans, expensive and ill-suited for unpredictable environments. Nature, in contrast, uses soft materials like muscle and skin, and has produced organisms that drastically outperform robots in terms of agility, dexterity, and adaptability. To create a new generation of lifelike robots that match the vast capabilities of biological systems, we need to develop actuators that replicate the astonishing all-around actuation performance of muscle. Hydraulically Amplified Self-healing ELectrostatic (HASEL) transducers are a new class of self-sensing, high-performance muscle-mimetic actuators, which are electrically driven and match or exceed most performance metrics of biological muscle; modeling results reveal rich underlying materials science to be further explored, and they lay out a roadmap towards HASELs with drastically improve performance, far surpassing both biological muscle and traditional electromagnetic motors. This talk gives an overview over the latest research results and commercialization efforts.