Keynote Lecture: Robot Skill Learning and Engineering

Abstract: With the advent of the large language models (LLMs), “end-to-end” large robot action models begin to blossom in very recent years with enormous enthusiasm for making humanoids and other robots intelligent. Initial results of recent advances seem promising, and major collaborative efforts are underway to collect demonstration data. For industrial automation applications, such as electronics assembly, the challenge is to enable robots with essential assembly skills in addition to the intelligence needed to deal with varying environmental conditions and dynamic obstacles on a factory floor. For robot skill acquisition, I argue that we need systematic approaches to integrate engineering modularity with learning, rather than a monolithic framework with the extensive and time-intensive online learning processes.
        Humans can effortlessly complete complex assembly tasks, relying on their acute visual and tactile sensing and skill learning abilities. Naturally, it is essential to endow robots with similar visual and tactile perception abilities as humans. This multi-modal perception necessitates robots to learn basic assembly skills and generalize these skills to other relevant situations to achieve comparable results. I’ll present a robot system for insertion tasks in unstructured environments, which integrates learning-based trajectory planning, passive interaction control, and a vision-based tactile sensor-guided alignment algorithm. The system permits a data-efficient training pipeline adept at learning from a limited set of demonstrations to generate human-like insertion trajectories, affirming the robustness and reliability of our approach.

Biography: Michael Yu Wang is a Chair Professor and the Founding Dean of the School of Engineering of the Great Bay University, China. He has served on the engineering faculty at University of Maryland, Chinese University of Hong Kong, National University of Singapore, Hong Kong University of Science and Technology, and Monash University. He has numerous professional honors–Kayamori Best Paper Award of 2001 IEEE International Conference on Robotics and Automation, the Compliant Mechanisms Award-Theory of ASME 31st Mechanisms and Robotics Conference in 2007, Research Excellence Award (2008) of CUHK, and ASME Design Automation Award (2013). He was the Editor-in-Chief of IEEE Trans. on Automation Science and Engineering, and served as an Associate Editor of IEEE Trans. on Robotics and Automation and ASME Journal of Manufacturing Science and Engineering. He is a Fellow of ASME, HKIE and IEEE. He received his Ph.D. degree from Carnegie Mellon University.

Keynote Lecture: Integrated experimental-numerical-virtual modelling technique

Abstract: Experimental techniques, involving the physical testing of materials, structures, or systems under controlled conditions, are fundamental for acquiring empirical data and validating theoretical models. Numerical simulations, with their sophisticated algorithms and capabilities, are extensively used for analysis, optimization, and safety assessment of materials and structures. Concurrently, virtual modelling techniques, notably those enhanced by recent advancements such as Extended Support Vector Regression (X-SVR), enable extensive testing and optimization within a virtual environment. These virtual techniques offer significant advantages in reducing the costs and time associated with physical experiments, particularly for scenarios that are financially prohibitive or impractical to simulate physically.
      This study presents a comprehensive framework that integrates experimental, numerical, and virtual modelling techniques. The framework is designed to leverage the full potential of machine learning technologies and computational mechanics, utilizing experimental data to address complex engineering challenges in a practical and effective manner. The research focuses on two critical engineering issues: stochastic response of composite structures under low/high velocity impacts and inverse-based stochastic prediction of structural elastoplastic behaviour.
      The integration of experimental, numerical, and virtual techniques within this framework demonstrates its potential to provide a more robust, efficient, and thorough approach to research and development. By leveraging the strengths of each method, this integrated approach enhances the accuracy of results and optimizes the efficiency of the design and analysis processes.

Biography: Professor Wei Gao has been working in the School of Civil and Environmental Engineering at the University of New South Wales (UNSW), Australia since 2008. He obtained his PhD in Mechanical Manufacturing and Automation from Xidian University, China in 2003. He worked at UNSW as a Vice-Chancellor’s postdoctoral research fellow from 2004-2007 and at University of Technology Sydney as a Chancellor’s research fellow from 2007-2008. He is currently an Associate Dean International of Faculty of Engineering at UNSW. Professor Gao works in challenging research areas bringing together computational mechanics, stochastic mechanics, mechanical and structural engineering. He has made profound contributions to computation uncertainty/stochastic mechanics, AI and machine learning in material/structural analysis and design, stochastic structural analysis and safety assessment in CAD/CAE. He has published over 300 papers including over 240 articles on international prestigious journals. His research outcomes have strong impact on the areas and have been cited frequently by researchers worldwide. In the last 15 years, he has secured 10 Australian Research Council Projects and has developed strong collaborations with industries to deliver the research outcomes to engineering practice. Professor Wei Gao is serving as an Associate Editor for Engineering Structures, a Subject Editor for Applied Mathematical Modelling, an Editorial Advisory Board Member for journals Nanotechnology Reviews, Machines, and Computer Methods in Applied Mechanics and Engineering.

Keynote Lecture: Making Functional “Hard” Materials by Soft Matter

Abstract: In this talk, I will not only talk about making the functional materials patterns from droplet array, but also our electrorheological elastomers can become hard reversibly.
      Combining droplet evaporative self-assembling, we can fabricate the micro/nanoscale all-inorganic perovskite single-crystal or thin-film arrays in one-step. Our method can adjust the pattern size and dewetting process and can be used to easily tune the perovskite crystal size position, with versatility in fabricating perovskite arrays in wafer scale. We studied systematically the split-ring topography from wettability, evaporative assembly to optoelectronics. We simultaneously fabricated the split-ring lyophilic patterns and electrode arrays using a dual-function laser etching technique. Compared to the fully lyophilic and square ring patterns, our split-ring pattern can capture 82% of precursor solution while the deposition area was reduced to 38%. This scheme not only assisted the highly efficient directional transportation of liquid, high-throughput fabrication of perovskite arrays and high compactness of perovskite film, but also simplified the preparation process and reduced the cost of the devices.
      Smart electrorheological (ER) materials are responsive colloids or suspensions whose viscosity or elastic properties can be reversibly tuned by an external electric field. They include ER fluids and ER elastomers. ER fluids are suspensions that dielectric particles suspended in a nonconducting liquid. ERF can change from liquid state to solid state reversibly and abruptly after electrical field is applied. ER elastomers are suspensions that dielectric particles suspended in a nonconducting elastomer. We synthesized the giant ER elastomer, and the settlement problem was solved totally. Its "soft and hard" state could be tuned by 100 times, ranking the best among the reported EREs. We succeed apply ERE in the smart table tennis racket.

Biography: Prof. Dr. Jinbo WU is a professor of Shenzhen MSU-BIT University and Adjunct Professor of Shanghai University. He was selected into Shanghai Pujiang Talent Program in 2016 and received the First Prize of Chongqing Science and Technology Progress Award in 2021. He serves as a member for Micro/nano-fluidic Technology Committee and Bio-MEMS Committee of the Chinese Society of Micro-nano Technology, the Electro-rheological and Magneto-rheological Committee in the Chinese Society of Theoretical and Applied Mechanics and Youth Editorial Board of Micro-Nano Letter. As a Principal Investigator, he supervised the National Natural Science Foundation of China, Shanghai Science and Technology Commission Project, Hong Kong Innovation and Technology Support Program, Hong Kong University of Science and Technology Overseas Research Fund. His research interests focus on soft materials, "Making the Functional Hard Materials from Soft Matter". He has published more than 70 papers in well-known SCI journals such as Nano-Micro Letters, Small, Engineering, etc., with more than 4000 citations. He already published one book (Functional and Intelligent Characteristics of Soft Matter and Its Applications, China Science Publishing) and 15 granted patents. His research findings attracted considerable attention and he delivered 40 invited talks in national and international conferences.

Keynote Lecture: Electromagnetic Regulation and Microwave Absorption Property of Dielectric-type Materials

Abstract: The research on high performance electromagnetic wave absorbing materials is a hot issue in the field of stealth technology. Dielectric-type materials usually possess unique electronic structure, adjustable electromagnetic parameters, and wide energy-band, which show potential application prospects in electromagnetic wave absorption area. To develop practical semiconductor nanomaterials with high intrinsic conductivity, wide-frequency band and high efficiency absorbing properties, the author focuses on solvothermal route preparation of dielectric-type materials, and employs surfactant-assisted approach to regulate the morphology and structure of dielectric-type nanomaterials, as well as to optimize the surface defects and electromagnetic properties. By introducing dielectric and magnetic multi-gradient dielectric/magnetic heterostructures, the impedance matching and attenuation performance were optimized, the enhancement mechanism of the microwave-absorbing performance was revealed, and the regulation strategies for the impedance match, dielectric properties and magnetic properties of the heterostructures were established, thus improving the reflection loss and broadening the effective wave-absorbing frequency band of materials. Currently, a series of breakthroughs including electromagnetic regulation, structural related interaction mechanism, absorbing performance, etc., have been made for VIB and VB group TMSs, multiple-metal MOFs, and high-entropy ceramics which provide new ideas for design of the new generation of low-frequency, wide-band and high-efficiency microwave absorbing materials, and lay scientific and technical foundation for the development of advanced stealth materials.

Biography: Dr. Junye Cheng currently works as Associate Professor at Shenzhen MSU-BIT University. He was honorably supported by the talent schemes “Pearl River Young Talents of Guangdong Province”, “Shenzhen high-level Talents”, “Leading Talents of Longgang District of Shenzhen”. He got his PhD degree from City University of Hong Kong in 2019. He is mainly engaged in the new nanomaterials design and their application in energy and environment related research. He especially focuses on the two-dimensional metal organic framework, ultra-thin metal sulfide nanosheets, high entropy materials and their controllable preparation of hybrid heterostructures for electrochemical energy storage devices and electromagnetic wave absorption. He has published more than 80 high-level academic papers in Adv. Mater., Electrochem. Energy Reviews, Adv. Funct. Mater., Adv. Energy. Mater., Nano-Micro Lett., Small., Ecomat etc with high citation >6900 times, and a h-index: 48 for him (Google Scholar data), 15 of which were listed as ESI papers with high citation. He was elected as one of the top 2% of the world's top scientists in 2024. As the project leader, he presided National Natural Science Foundation of China project (2), Basic Research Project of Guangdong Province (2) etc. He was ever awarded 2023 ESI TOP Article AWARD by Nano-Mirco Letters journal, 2022 Best Paper Award by IOP publisher, the Most Influenced Paper Award in 2021 by Science Bulletin journal, the Best Poster Award of the European Materials Research Society (EMRS) in 2019, and the Outstanding Academic Research Award of City University of Hong Kong for three consecutive years. He also serves as young editorial board member of Nano-Micro Letters, Nano Research Energy and Environmental Science and Ecotechnology, guest editor for Frontiers in Materials and Nanomaterials, member for American Materials Research Society, invited member for American Chemistry Society. He has been invited as keynote speaker or invited speaker in international conferences many times. He also served as chairman or session chairman in the international conferences many times.