Mei Li, Professor
China University of Geosciences (Beijing) , China
Biography: Mei Li, female, born in July 1969, Ph.D., Professor and Doctoral Supervisor. Research areas: Artificial Intelligence, Self-evolving Agents, Computer Vision, Knowledge Graphs, Internet of Things and Big Data Technologies. She has been working at China University of Geosciences (Beijing) since 2003, mainly engaged in teaching and research on signal and information processing.
Speech Title: Psychological and Emotional Companion Agent Based on MyAI
Abstract: This talk introduces a psychological and emotional companion agent developed based on MyAI. Rooted in person-centered therapy, positive psychology and emotional coaching, this AI agent provides lightweight daily emotional support. It practices non-judgmental listening, captures emotional cues and helps users reflect on inner feelings in easy-to-understand language. Please note that it is merely an auxiliary tool and cannot replace licensed counselors. Those with serious mental problems should seek offline professional help. This work explores how AI can deliver convenient emotional accompaniment for parents and the general public.
Zhifeng Huang, Associate Professor
Guangdong University of Technology, China
Biography: Dr. Zhifeng Huang is an Associate Professor, Doctoral Supervisor, and Deputy Director of the Department of Automatic Control at the School of Automation, Guangdong University of Technology, and also a Visiting Researcher at the The Research into Artifacts, Center for Engineering of the University of Tokyo; he received his B.Eng. from South China University of Technology (2007), M.Eng. from Harbin Institute of Technology (2010), and Ph.D. from the University of Tokyo (2014). His research focuses on mechanism design and control theory for humanoid robots and applications of mobile manipulators, and he has published over 20 SCI/EI-indexed papers in leading journals and conferences such as IEEE Transactions on Mechatronics, IEEE Transactions on Learning Technologies, IEEE Robotics and Automation Letters, IROS, and ICRA, along with 12 authorized invention patents. As principal investigator, he has secured one National Natural Science Foundation project, three Guangdong Provincial Natural Science Foundation projects, one key R&D project for central state-owned enterprises (¥1.55 million), and ten industry collaborative projects, with cumulative research funding exceeding ¥3 million RMB. In 2018, he was honored with the title of "Innovation Hero" by the Guangzhou Municipal Publicity Department. His notable achievements include developing in 2017 the world's first bipedal robot based on a ducted‑fan propulsion system for large‑scale obstacle crossing, which set a world record among humanoid robots with a stepping distance reaching 147% of its own leg length; and in 2021, he designed the world's first flying bipedal robot using vector‑thrust control and successfully demonstrated controllable take‑off. This series of research has been widely reported by both domestic and international media, including IEEE Spectrum, the Daily Mail (UK), Guangdong Television, China Daily, Southern Daily, and China News Service.
Speech Title: Flying Humanoid Robots: Structural Innovation, Control, and Experimental Validation
Abstract: Aerial locomotion serves as an efficient means for humanoid robots to rapidly approach targets and execute missions, offering significant application value in disaster emergency response, high-altitude operations, and isolated-area rescue. This keynote reviews the state-of-the-art development of flying humanoid robots and multi-link deformable aerial vehicles, and highlights the speaker's latest research achievements and technical breakthroughs in this field. The presentation is organized as follows: (1) innovative structural designs of ducted-fan-propelled flying humanoid systems; (2) a model predictive control (MPC)-based full-body control framework for stable aerial maneuvering; and (3) systematic experimental validations on physical prototypes, including aerial trajectory tracking, autonomous take-off and landing, and in-flight deformation.
Cunman Liang, Associate Professor
Tianjin University, China
Biography: Dr. Cunman Liang is an Associate professor at the School of Mechanical Engineering, Tianjin University, Deputy Director of the International Joint Research Center for Micro/Nano Manufacturing Technology and Equipment, and a recipient of the Young Talent Recruitment Program launched by the Ministry of Education of China. He earned his Bachelor’s, Master’s and Doctoral degrees in Mechanical Engineering from Tianjin University in 2012, 2015 and 2020 respectively. From 2017 to 2018, he completed joint doctoral training at Northwestern University in the United States under the guidance of Professor John A. Rogers. From 2020 to 2023, he carried out postdoctoral research at The Chinese University of Hong Kong and the Hong Kong Centre for Cardio-Cerebrovascular Health Engineering (COCHE). Dr. Liang has presided over a number of research projects, including programs funded by the National Natural Science Foundation of China, the Overseas Talent Recruitment Special Project of the Ministry of Education and open projects of State Key Laboratories. He has published over 50 academic papers in journals such as Science Robotics and Nature Communications, alongside 2 book chapters. His work has accumulated more than 3,000 citations on Google Scholar with an H-index of 25. He also owns over 10 authorized national patents and has filed 2 U.S. patents. He has received numerous academic accolades, including the 2016 IEEE 3M-NANO Best Student Paper Award, the 2024 IEEE 3M-NANO Best Conference Paper Nomination, the Excellent Youth Editorial Board Member Award of SmartBot and the 2024 Pioneer Award from the School of Mechanical Engineering, Tianjin University. Currently serving as a Youth Editorial Board Member of four journals, including Soft Science and SmartBot, he was invited to deliver a keynote speech at the 2023 Annual Conference of the Chinese Society of Micro-Nano Technology. His research mainly concentrates on micro/nano robotics and micro/nano sensing technologies, and his outstanding research outcomes have been featured by more than ten mainstream media platforms including Sina, Sohu, China Science and Technology Network.
Speech Title: Bioinspired Magnetically Actuated Miniature Soft Robots
Abstract: Miniature soft robots are robotic systems with characteristic dimensions at the millimeter or micrometer scale. Among them, magnetically actuated miniature soft robots have attracted considerable research interest because of their untethered actuation, flexible control, and strong environmental adaptability. However, achieving versatile and robust locomotion in confined, unstructured, and complex environments remains a major challenge. This talk presents two complementary studies that address this challenge through bioinspired structural design and tailored magnetic actuation strategies. The first study introduces a miniature amphibious soft robot capable of crawling, rolling, jumping, and swimming. A polyhedral deployable structure with flexible joints concentrates deformation and elastic energy storage at designated compliant regions, thereby improving the efficiency of magnetic actuation. Multi-objective optimization of the robot’s structural dimensions balances the competing performance requirements of its four locomotion modes. In addition, vision-based closed-loop control enables precise planar trajectory tracking. Experiments in complex obstacle environments and an ex vivo porcine stomach demonstrate the robot’s ability to navigate confined and unstructured environments by transitioning between different locomotion modes. The second study focuses on robust swimming under fluid disturbances. Inspired by the autorotation of Dipterocarpus alatus seeds and the pulsatile propulsion of jellyfish, the robot combines inclined magnetized legs with a flexible polyhedral deployable body to generate pulsatile thrust and stabilizing self-rotation. A resonance-based actuation strategy increases the maximum swimming speed to 43.7 mm/s, while the self-rotational angular velocity reaches 4.7 rad/s. Analysis of the coupling between propulsion and self-rotation provides guidance for selecting suitable actuation conditions that balance swimming speed and stability. When subjected to a lateral water flow of 150 mm/s, the robot recovers its stability within 1.3 s. It also demonstrates directional navigation and targeted pH sensing in a confined underwater space. Together, these studies show how the coordinated design of bioinspired structures and magnetic actuation strategies can enable both versatile locomotion across different environments and robust movement under external disturbances, offering new possibilities for miniature soft robots in complex terrestrial and aquatic environments.
Heng Wang, Associate Professor
South China University of Technology, China
Biography: Dr. Heng Wang is an Associate Professor with the Shien-Ming Wu School of Intelligent Engineering, South China University of Technology, Guangzhou, China, and the director of SCUT-Shanghai Huayi Tech joint Medical Virtual Reality Lab. He received the bachelor’s degree from Xi’an Jiaotong University, China, in 2016 and received the Ph.D. degree in Mechanical Engineering from University of Minnesota, Twin Cities, USA, in 2021. His research interests include magnetic motion tracking for surgical robot navigation and embodied intelligence training, magnetically actuated capsule robots and continuum robots, medical robotics, and medical VR/AR and human-machine interaction. Dr. Wang is a principal investigator of multiple research projects including NSFC project, Guangdong and Guangzhou science and technology projects, and numerous industry-funded projects. Dr. Wang was granted a US patent and three China patents, and published more than 20 research papers in renowned journals and conferences such as Nature Communications/TRO/TMECH/TIM/MSSP/RAL/ACC, etc. He received the Doctoral Dissertation Fellowship from University of Minnesota in 2020, the honor of Editor’s Highlight paper (Best 50 papers) in “Applied Physics and Mathematics” in Nature Communications, ASME Best Mechatronics Paper Award in 2020, Best Application Paper Award in the 2022 IEEE International Conference on Digital Twins and Parallel Intelligence.
Speech Title: Versatile Magnetic Navigation Systems for Precise Medical Robotics
Abstract: Miniature medical robots, e.g., capsule robots and transluminal flexible robots, have been increasingly developed in recent years for minimally invasive medicine due to their small size and easy access to narrow lumens. However, there exists significant technical challenges in precise navigation and dexterous actuation of such miniature robots through tortuous and complicated intracorporeal lumens such as gastrointestinal tracts, bronchi, and blood vessels. Magnetic navigation systems are advantageous for intracorporeal medical robots because the magnetic field can permeate the human body safely and achieve remote sensing and actuation. In this talk, diverse magnetic sensing and actuation systems will be discussed for wireless, autonomous, and precise navigation of miniature medical robots in minimally invasive medicine. First, magnetic pose tracking systems with novel magnetic materials, configurations of magnetic sources, estimation algorithms, and field control strategies are discussed. These innovations give magnetic tracking systems more degrees of freedom, better accuracy, larger coverage, and more uniform sensitivity. A novel magnetic shape sensing method based on magnetic elastomers is also discussed for continuum medical robots. Second, magnetic actuation systems are developed for full 6-DoF dexterous manipulation of miniature capsule robots. The added DoF comes from either the use of anisotropic soft magnet as the magnetic actuation element or the skew configuration of the Helmholtz-Maxwell coils. Finally, an open-source simulator for magnetic robots is developed to facilitate research and development of such robots. In this seminar, the versatility of magnetic systems for medical robots is demonstrated by their diverse configurations, functions and applications.
Lin Lin, Associate Professor
Southern University of Science and Technology (SUSTech), China
Biography: Dr. Lin Lin is an Associate Professor and Ph.D. Supervisor at the Southern University of Science and Technology (SUSTech). She received her Ph.D. degree from the Department of Mechanical Engineering at The University of Hong Kong. During her doctoral studies, she worked as a visiting research student at the University of Adelaide (Australia), King's College London (UK), and Hokkaido University (Japan). She subsequently held postdoctoral fellow positions at City University of Hong Kong, The University of Hong Kong, and the Multi-Scale Medical Robotics Center. Her primary research interests include swarm intelligence networks, fully actuated system methods, lightweight control, and medical robotics. She has published over 20 papers as the first author in prestigious international journals such as IEEE Transactions, Automatica, and SICON. Dr. Lin was a receipt of the Excellent Young Scholarship of the National Natural Science of Foundation of China (Overseas), the Humboldt Research Fellowship, IETI PhD Fellowship Award, and Hong Kong Young Scientist Award (Honorable mention in Engineering Science).
Speech Title: Analysis, Synthesis, and Learning of Finite-Valued Networks
Abstract: The talk will discuss the analysis, synthesis and learning problems for several classes of network-inspired finite-valued systems, including finite-field networks, Boolean networks, logical dynamic systems, and Markov chains. They are effective models to describe agents operating under realistic circumstances with limited capacities for storing, processing and transmitting information. To tackle the challenges posed by high computational complexity and large memory storage, the present investigation is concerned with several approaches, such as establishing a quotient/bisimulation-based scale reduction method, designing optimal triggering control strategy, exploring network structure information, leveraging data structure storage, employing graph-theoretical algorithms, and incorporating reinforcement learning techniques.
Peng Chen, Associate Professor
Shantou University, China
Biography: Dr. Peng Chen is an Associate Professor and Master’s Supervisor at Shantou University, recognized as a Shantou High-Level Talent and selected for the university’s “Outstanding Talents Program.” He received his Ph.D. in Engineering from the University of Electronic Science and Technology of China (UESTC) in 2020, under the supervision of Prof. Ming J. Zuo — IEEE Fellow and Fellow of the Canadian Academy of Engineering — with joint doctoral training at KU Leuven (Belgium) and a visiting research appointment at the University of Pretoria (South Africa). He is an IEEE member, an external peer reviewer for the National Natural Science Foundation of China (NSFC) and serves on multiple technical committees of the Chinese Society for Vibration Engineering, including the Signal Processing Branch, Dynamic Testing Professional Committee, and Rotor Dynamics Branch. As a principal investigator, he has led over 10 research projects — including an NSFC Young Scientists Fund project rated “Excellent” upon completion and an AVIC Key R&D sub-project — while also contributing to six major national-level initiatives. He has published over 50 SCI-indexed journal articles as first or corresponding author in premier venues such as MECH SYST SIGNAL PR、 EXPERT SYST APPL 、KNOWL BASED SYST、 ENG APPL ARTIF INTELL 、IEEE INTERNET THINGS J、 IEEE T INSTRUM MEAS、 IEEE T RELIAB, among others (featuring two Highly Cited/Hot Papers), and holds six granted invention patents. He has served as session chair and delivered invited talks at international conferences including UNIfied-2026-SMMI, IMCC 2025, and TEPEN2024-IWFDP, and actively contributes to the academic community as a reviewer and guest editor for multiple prestigious international journals.
Speech Title: Collaborative Representation and Decoupling for Fault Diagnosis of Robotic and Mechatronic Systems under Transient Interference and Multi-Source Coupling
Abstract: Transient interference and multi-source coupling obscure weak fault signatures and complicate fault diagnosis in robotic and mechatronic systems. Superposed responses from motors, bearings, and transmission components can mask fault-induced impulses and hinder source separation. This talk presents collaborative representation and decoupling methods for extracting interpretable diagnostic information from coupled responses. Adaptive transient detection and signal reconstruction suppress interference while preserving fault-related periodicity. Walsh–Fourier representations and cross-correlation heat maps support joint feature characterization, while target-guided multiband demodulation and confidence-weighted fusion integrate diagnostic evidence across frequency bands. Particular attention is given to distinguishing fault-induced impacts from external disturbances and separating overlapping source contributions. Applications to robotic manipulators and mechanical transmission systems illustrate the relevance of these approaches to fault discrimination and condition assessment. Future directions include learning with limited labeled data, multimodal sensing, and efficient online implementation. The discussion connects these methods with condition monitoring and predictive maintenance requirements in industrial robotics, automated production lines, and precision manufacturing equipment.
Yunquan Li, Associate Professor
South China University of Technology, China
Biography: Yunquan Li, Associate Professor and Doctoral Supervisor at the Shien-Ming Wu School of Intelligent Engineering, South China University of Technology. He received his Ph.D. in Mechanical Engineering from the University of Hong Kong and completed postdoctoral research at the HKU Faculty of Medicine. His research interests include embodied intelligent robots, dexterous end-effectors, high-sensitivity tactile sensors, bio-inspired robotics, and mobile robotics. He has published 59 papers with over 1,300 Google Scholar citations and an h-index of 19. His work has appeared in IEEE TIE, T-MECH, Soft Robotics, and other leading journals. He holds 4 invention patents. He has led the NSFC project, several provincial and municipal projects, and industry-university collaborations with Midea Group and SIAT. He was selected for the Guangzhou Young Top-Talent Program and is a core member of the Guangzhou Embodied Intelligent Robot Innovation and Entrepreneurship Team. He received the Excellence Award in the Intelligent Robot Championship at the National Disruptive Technology Innovation Competition and the Wiley China Top-Cited Author Award.
Speech Title: Low-Cost and Highly Sensitive Optical Tactile Sensors for Robotic Force Perception
Abstract: This report presents a comprehensive overview of our recent research on low-cost, highly sensitive flexible optical tactile sensors for intelligent robots. The presentation will first introduce the background and significance of tactile perception in human-robot interaction and complex operations. It will then cover two primary technical approaches developed in our lab: 1) A particle-jamming gripper integrated with a stretchable optical waveguide skin for large-area shape-adaptive grasping and slip detection; 2) A compact optical blocking structure-based tactile sensor for high-sensitivity 1D/3D force detection. The report will also discuss how machine learning algorithms (such as GRU and SVM) are utilized to achieve accurate 3D force decoupling, contact position recognition, and grasping posture classification. Finally, practical applications in robotic arms, including adaptive grasping, hardness/texture detection, and multi-axis teleoperation, will be demonstrated.