|Fuchun Sun||Robot skill learning, transfer and enhancement for dexterous operation applications||Saturday July 3, 08:50-9:30|
|Yaochu Jin||Data-Driven Evolutionary Optimization||Saturday July 3, 09:30-10:10|
|Xin Zhao||Precise Robotic Enucleation to Increase Success Rate of Cloned Animal||Saturday July 3, 10:10-10：50|
|TAN Kay Chen||Recent Advances in Evolutionary Transfer Optimization||Saturday July 3, 10:50-11：30|
|Daoyi Dong||Improved reinforcement learning with applications in robotics, games and quantum control||Saturday July 3, 11:30-12:10|
|Mark Yim||Symmetry in Underactuated Robots||Sunday July 4, 08:50-09:30|
|Junzhi Yu||Bioinspired Underwater Robots and Their Applications||Sunday July 4, 09:30-10：10|
|Zhidong Wang||Cooperative Robot Control with Uncertainties: from designing human-robot cooperation to mapping human motion behavior, and manipulating micro/nano objects||Sunday July 4, 10:10-10:50|
|Long Cheng||Recent Advances on Hand Rehabilitation Robots for Post-Stroke Patients||Sunday July 4, 10:50-11：30|
Yaochu Jin, Professor, University of Surrey, UK
Data-Driven Evolutionary Optimization
Abstract: Many real-world optimization problems are data-driven, where no analytical objective or constraint functions are available and time-consuming numerical simulations or expensive physical experiments must be done to perform optimization. Over the past two decades, data-driven surrogate-assisted evolutionary optimization has attracted increasing interest both in industry and academia. This talk will introduce the basic ideas of data-driven surrogate-assisted evolutionary optimization, present main advances , in particular in handling high-dimensional and many-objective problems, and discuss the remaining challenges. Applications of data-driven optimization, ranging from aerodynamic design optimization to deep neural architecture search will be given.
Biography: Professor Jin received the BSc, MSc, and PhD degrees from Zhejiang University, Hangzhou, China, in 1988, 1991, and 1996, respectively, and the Dr.-Ing. degree from Ruhr University Bochum, Germany, in 2001.
He is currently a Distinguished Chair, Professor in Computational Intelligence, Department of Computer Science, University of Surrey, Guildford, U.K., where he heads the Nature Inspired Computing and Engineering Group. He was a “Finland Distinguished Professor” of University of Jyvaskyla, Finland, a “Changjiang Distinguished Visiting Professor”, Northeastern University, China, and “Distinguished Visiting Scholar”, University of Technology Sydney, Australia. His main research interests include data-driven surrogate-assisted evolutionary optimization, trustworthy machine learning, multi-objective evolutionary learning, swarm robotics, and evolutionary developmental systems.
Dr Jin is presently the Editor-in-Chief of the IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS and the Editor-in-Chief of Complex & Intelligent Systems. He was an IEEE Distinguished Lecturer, and Vice President of the IEEE Computational Intelligence Society. He is the recipient of the 2018 and 2020 IEEE Transactions on Evolutionary Computation Outstanding Paper Award, the 2014, 2016, and 2019 IEEE Computational Intelligence Magazine Outstanding Paper Award, and the Best Paper Award of the 2010 IEEE Symposium on Computational Intelligence in Bioinformatics and Computational Biology. He is recognized as a Highly Cited Researcher 2019 and 2020 by the Web of Science Group. He is a Fellow of IEEE.
He obtained the BSc, MSc and PhD degrees from Zhejiang University, Hangzhou, China and the Dr.-Ing. from Ruhr-University Bochum, Germany. Before joining Surrey in 2010, he was a Principal Scientist with Honda Research Institute Europe, Germany. He did postdoctoral research with the Industrial Engineering Department, Rutgers, the State University of New Jersey, USA from 1998 to 1999. he was an Assistant and Associate Professor with the Electrical Engineering Department, Zhejiang University, Hangzhou, China from 1992 to 1996.
Zhidong Wang, Professor, Chiba Institute of Technology, Japan
Cooperative Robot Control with Uncertainties: from designing human-robot cooperation to mapping human motion behavior, and manipulating micro/nano objects.
Abstract: Controlling multiple autonomous robots and human-robot system in coordination are interesting and challenging research topics, especially to the mobile robot system without explicit inter-robot communication. In this talk, two robot systems having physical interactions among humans and robots will be introduced. In these systems, each robot is controlled as if it has a specified impedance dynamics, and a leader-follower type control algorithm is incorporated for estimating the human/leader robot desired motion based on the intentional force/moment applied by the human and the information of an environment. A Dance robot system is mainly designed for human intention estimation and skill evaluation as a whole-body motion with knowledge based system and dynamic interaction. These examples will inspire possible applications of the human-robot interaction in near future.
Recently, we also proposed a concept and architecture of Human Motion Map by representing extracted human behavior in the human living space as a map, by using human state estimation function and mapping function of SLAM. The concept is implemented in a mobile robot system as a high dimensional map structure with multi-layer representing some basic motions of human being in particular place in the map, which is generated from individual observations of hundred’s experiments. A motion feature descriptor is developed based on Human Motion Map for representing various walking behaviors in indoor environments and applying machine learning architectures. Furthermore, some recent results on caging based cooperative micro-bubble robot control for living cells microassembly, nano scale SLAM based localization with local-scan method, and nano-partical manipulation with nano-hand strategy will be presented for coping with significant uncertainties in cooperative micro/nano object handling.
Biography: ZhiDong Wang received his Bachelor of Engineering from Beihang University, China in 1987, and received his Master degree and Ph.D in Engineering from Tohoku University, Japan in 1992 and 1995 respectively. From 1995, he joined the Advanced Robotics Laboratory and later the Intelligent Robotics Laboratory at Tohoku University as an assistant and associate professor respectively. From 2006, he joined the Department of Advanced Robotics, Chiba Institute of Technology, and is currently professor and head of Biomimetic Systems Lab. at CIT, Japan.
Dr. Wang has published numerous journal and conference articles. He and his colleague received several best paper awards including the 2014 ROBIO Best Paper in Robotics Award, the JSME Award for best paper in 2005, and 2005 IROS Cyberbotics Award for Best Paper in Experimental Robotics, 2019 ROBIO Best Paper in Robotics Award. He served several academic meetings and was General Chair of ROBIO2011, Cyber2014, a Program Chair of ROBIO2007, Nanomed2016, and Program Co-Chair of ICRA2011, IROS2013. He will serve General Chair of ICRA2024 at Yokohama, Japan. Currently, he is serving the Vice President of ESPB and Associate Vice President of CAB of IEEE Robotics and Automation Society. His main research interests are human-robot interaction, distributed robotics, nano-manipulation, and application of cooperative robotics.
Fuchun Sun, Professor, Tsinghua University, China
Robot skill learning, transfer and enhancement for dexterous operation applications
Abstract: The development of artificial intelligence is gradually changing from scene intelligence dominated by open-loop learning to behavioral intelligence dominated by closed-loop learning. Behavioral intelligence not only emphasizes the perception and processing of simulated human brain information, but also emphasizes brain body co-development, i.e. perception and behavior as two physical processes coordinate with each other under the command of brain cognitive body, to solve the dynamic, interactive and adaptive problems of behavior learning in complex tasks. As the core of behavioral intelligence, skill learning for robot manipulations is a difficult and hot issue in current research field. In view of the problems that the existing skill learning methods do not make full use of the expert demonstrations and cannot achieve efficient policy learning, and the imitation learning algorithm is sensitive to the expert preference characteristics and the local manipulation space, this report introduces the theoretical and technical achievements in perception fusion of visual, tactile and acoustic information, imitation learning, skill transfer and enhancement of robot manipulation. Then, the application of skill transfer learning and enhancement technologies in operation of UAVs and robot dexterous manipulation will be introduced. Finally, the development trend of robot manipulation skill learning will be discussed.
Biography: Dr. Fuchun Sun is professor of Department of Computer Science and Technology, President of Academic Committee of the Department, Tsinghua University, and deputy director of State Key Lab. of Intelligent Technology & Systems, Beijing, China. He serves as Vice Chairman of Chinese Association for Artificial Intelligence and Executive Director of Chinese Association for Automation. His research interests include robotic perception and skill learning，Cross-modal Learning and intelligent control. He has won the Champion of Autonomous Grasp Challenges in IROS2016 and IROS 2019. He is elected as IEEE Fellow and CAAI Fellow in 2019, CAA Fellow in 2020.
Dr. Sun is the recipient of the excellent Doctoral Dissertation Prize of China in 2000 by MOE of China and the Choon-Gang Academic Award by Korea in 2003, and was recognized as a Distinguished Young Scholar in 2006 by the Natural Science Foundation of China. He served as the EIC of the Journal of Cognitive Computation and Systems, and associated editors of IEEE Trans. on Neural Networks and Learning Systems during 2006-2010, IEEE Trans. On Fuzzy Systems since 2011, IEEE Trans. on Cognitive and Development Systems since 2018 and IEEE Trans. on Systems, Man and Cybernetics: Systems since 2015.
Mark Yim,Professor, Asa Whitney Professor of Mechanical Engineering, Director of the GRASP Lab, University of Pennsylvania
Symmetry in Underactuated Robots
Abstract: Symmetry is often thought of as natural, desirable or elegant in many engineered systems. In robotics, it often leads to compact efficient control and computation. Underactuated robots effectively control more degrees of freedom than the number of actuators. This can lead to lower cost systems with interesting engineering puzzles to solve with interesting questions: Can you control a drone to fly in 3D space with just one motor? Can you make a robot gripper that has no motors? Can diff-drive be holonomic? The presented devices and systems taken as a whole result in general principles that guide cost-effective systems which all share one aspect – a lack of symmetry.
Biography: Mark Yim is the Asa Whitney Professor of Mechanical Engineering in the School of Engineering and Applied Science. Yim is the director of the GRASP Lab, the oldest robotics research laboratory in the country established in 1980. His research group designs and builds a variety of electromechanical hardware. Demonstrations range from a humanoid robot on display at the Philadelphia Museum of Art to transforming robots that can change their shape to the smallest self-powered flying robot in the world. His other research interests include product design, robotic performance art, novel locomotion, low-cost manipulation, in the search and rescue as well as healthcare applications. Honors include the Lindback Award for Distinguished Teaching (UPenn’s highest teaching honor); induction to MIT’s TR100 in 1999; induction to the National Academy of Inventors. He has over 200 publications and over 50 patents issued (perhaps the most prominent patents are related to the video game vibration control which resulted in over US$100 million in litigation and settlements). He has started three companies, one in robotics and one medical device company making a steerable needle and one focused on thermal storage.
Daoyi Dong, Professor, University of New South Wales, Australia
Improved reinforcement learning with applications in robotics, games and quantum control.
Abstract: Reinforcement learning (RL) addresses the problem of how an autonomous active agent can learn to approximate an optimal behavioral strategy while interacting with its environment. It has been widely applied in various areas including artificial intelligence, control engineering, operations research and robotics. In this talk, we will introduce several improved reinforcement learning algorithms which were developed by my collaborators and myself. These algorithms include incremental reinforcement learning, quantum reinforcement learning, quantum-inspired deep reinforcement learning. We will also demonstrate several applications of these improved reinforcement learning algorithms to robotics, games and quantum control.
Biography: Professor Dong is currently a Scientia Associate Professor at the University of New South Wales, Canberra, Australia, and he is also an Alexander von Humboldt Fellow. He was with the Chinese Academy of Sciences and with the Zhejiang University. He had visiting positions at Princeton University, USA, RIKEN, Japan and the University of Hong Kong, Hong Kong, and University of Duisburg-Essen, Germany. He received a B.E. degree in automatic control and a Ph.D. degree in engineering from the University of Science and Technology of China, in 2001 and 2006, respectively. His research interests include machine learning and quantum control. He was awarded an ACA Temasek Young Educator Award by the Asian Control Association and is a recipient of an International Collaboration Award, Discovery International Award and an Australian Post-Doctoral Fellowship from the Australian Research Council, and Humboldt Research Fellowship from Alexander von Humboldt Foundation in Germany. He serves as an Associate Editor of IEEE Transactions on Neural Networks and Learning Systems, and Technical Editor of IEEE/ASME Transactions on Mechatronics. He has also served as General Chair or Program Chair for several international conferences, and is currently Associate Vice-President and a Member-at-Large of Board of Governors, IEEE Systems, Man and Cybernetics Society. He has published 105 journal papers in leading journals including Nature Human Behaviour, Physical Review Letters, IEEE Transactions, and Automatica, and more than 50 conference paper. He has attracted a number of competitive grants with more than AU$2.8 million from Australia, USA, China and Germany.
TAN Kay Chen, Professor, Hong Kong Polytechnic University, China
Recent Advances in Evolutionary Transfer Optimization.
Abstract: It is known that the processes of learning and the transfer of what has been learned are central to humans in problem-solving. However, the study of optimization methodology which learns from the problem solved and transfer what have been learned to help problem-solving on unseen problems, has been under-explored in the context of evolutionary computation. This talk will touch upon the topic of evolutionary transfer optimization (ETO), which focuses on knowledge learning and transfer across problems for enhanced evolutionary optimization performance. It will first present an overview of ETO for problem-solving in evolutionary computation. It will then introduce our recent work on ETO for evolutionary multitasking which is an emerging search paradigm in the realm of evolutionary computation that conducts evolutionary search concurrently on multiple search spaces corresponding to different tasks or optimization problems. It will end with a discussion of future ETO research directions, covering various topics ranging from theoretical analysis to real-world complex applications.
Biography: Prof. Kay Chen Tan is currently a Chair Professor (Computational Intelligence) of the Department of Computing, the Hong Kong Polytechnic University. He has co-authored 7 books and published over 200 peer-reviewed journal articles.
Prof. Tan is currently the Vice-President (Publications) of IEEE Computational Intelligence Society, USA. He was the Editor-in-Chief of IEEE Transactions on Evolutionary Computation from 2015-2020 (IF: 11.169) and IEEE Computational Intelligence Magazine from 2010-2013 (IF: 9.083). Prof. Tan currently serves as an Associate Editor of various international journals, such as IEEE Transactions on Artificial Intelligence, IEEE Transactions on Cybernetics, and IEEE Transactions on Games.
Prof. Tan has been invited as a Plenary/Keynote speaker for many international conferences, including the 2020 IEEE World Congress on Computational Intelligence, the 2016 IEEE Symposium Series on Computational Intelligence, etc. He has served as an organizing committee Chair/Co-Chair for many international conferences, including the General Co-Chair of 2019 IEEE Congress on Evolutionary Computation, and the General Co-Chair of 2016 IEEE World Congress on Computational Intelligence, etc.
Prof. Tan has received a number of research awards, such as the 2020 IEEE Transactions on Cybernetics Outstanding Paper Awards, the 2019 IEEE Computational Intelligence Magazine Outstanding Paper Awards, the 2016 IEEE Transactions on Neural Networks and Learning Systems Outstanding Paper Awards, the 2012 Outstanding Early Career Award presented by the IEEE Computational Intelligence Society.
Xin Zhao, Chair Professor, Dean of Artificial Intelligence, Nankai University, Tianjin, China.
Precise Robotic Enucleation to Increase Success Rate of Cloned Animal.
Abstract: The somatic cell nuclear transfer (SCNT), also known as animal clone, is one of most complex and challenging cell manipulation tasks. The SCNT involves multiple manipulation procedures, such as oocyte rotation, penetration, enucleation, and somatic cell injection, and inevitably causes intracellular damage to recipient oocytes during manipulation, resulting in only around 1-2% of reconstructed embryos developed into live cloned animals. The low success rate has become the major obstacle to extensive applications of the SCNT. In this talk, the automated polar body detection and nuclei visualization techniques were developed to perform precise enucleation through reducing the amount of lost cytoplasm in enucleation. Then, a robotic SCNT system was established and applied to pig cloning. We did thousands of robotic SCNT operations and transferred 510 reconstructed embryos to 6 pigs, and obtained 17 cloned pigs at last. Compared to manual SCNT methods, the blastocyst rate of our system was improved from 10% to 21%, the clone success rate was improved from 1-2% to 3.3%.
Biography: Prof. Zhao received the B.S. degree from Nankai University, Tianjin, P.R.China, in 1991, the M.S. degree from Shenyang Institute of Automation, CAS, Shenyang, P.R.China, in 1994 , and the Ph.D. degree from Nankai University, in 1997, all in control theory and control engineering. He joined the faculty at Nankai University, Tianjin, P.R.China in 1997. He was a Visiting Professor in Center of Cell Control, Dept. of Mechanical & Aerospace Engineering, University of California at Los Angeles in 2009-2010. His research interests are in Mico-Nano Manipulation and System and Mathematical Biology. Prof. Zhao was the recipient of 1999 Excellent Professor Award, Nankai University, 2000 Inventory Prize, Tianjin Municipal Government, 2002 Excellent Professor Award of “College Key Teachers Fund”, Ministry of Education, 2002 Excellent Professor Award of “Baogang Fund” and 2007 Program for New Century Excellent Talents in University, Ministry of Education. His team was supported by High Level Innovation Team in Tianjin Special Support Plan for Talents Development and Tianjin Key Areas Innovation Team (2017). His team conducted the first batch of robotic-operated alive cloned animals around the world in 2017 and received the Award of China’s 10 Advancements in Intelligent Manufacturing Science and Technology in 2018.
Junzhi Yu, Professor, College of Engineering, Peking University.
Bioinspired Underwater Robots and Their Applications.
Abstract: Robotic fish, inspired by fish in nature, have drawn much attention in the last two decades. As an excellent research and experimental platform, robotic fish not only plays an important role in helping biologists to investigate the kinematic mechanism and hydrodynamic analyses, but also is employed by engineers to explore practical, versatile and flexible propulsive mechanisms since natural fish have acquired such surprised swimming skills characterized by high effectiveness, high maneuverability, and low noise. Since the first robotic fish, RoboTuna, was created at MIT in 1994, more and more robotic fish prototypes have been developed to explore the high efficiency and high maneuverability in fishlike swimming. In this talk, I will first introduce the main motion characteristics of real fish and summarize a general research technical route for the bioinspired robotic fish. Then, on the basis of our recent research achievements in biomimetic robotic fish and robotic dolphin, I will emphatically elaborate the analysis and control for high-efficiency and high-maneuverability motion of the robotic fish and robotic dolphin. Remarkably, acrobatic flips and leaps which are first implemented by the physical robots will also be detailed. In additional, some aquatic scenario related applications will be mentioned.
Biography: Junzhi Yu received the B.E. degree in safety engineering and M.E. degree in precision instruments and mechanology from the North University of China, Taiyuan, China, in 1998 and 2001, respectively, and the Ph.D. degree in control theory and control engineering from the Institute of Automation, Chinese Academy of Sciences, Beijing, China, in 2003. From 2004 to 2006, he was a Postdoctoral Research Fellow with the Center for Systems and Control, Peking University, Beijing. He was an Associate Professor with the Institute of Automation, Chinese Academy of Sciences in 2006, where he became a Full Professor in 2012. He was an AvH research fellow with the University of Hamburg, Germany, from September 2009 to September 2011. In 2018, he joined the College of Engineering, Peking University, as a Tenured Full Professor.
For his achievements in swimming robots, Dr. Yu received the Outstanding Young Investigator Award from the National Natural Science Foundation of China and a National Natural Science Award, China, in 2017. In 2020, he was elected Fellow of IEEE through the Robotics and Automation Society. He has authored or coauthored more than 100 peer-reviewed international journal papers and five monographs in the areas of bioinspired swimming robots, motion control, and visual perception. He serves or has served as an associate editor for IEEE Transactions on Robotics, IEEE/ASME Transactions on Mechatronics, Bioinspiration & Biomimetics, Journal of Bionic Engineering, etc. His research interests include intelligent robots, motion control, and intelligent mechatronic systems.
Long Cheng, Institute of Automation，Chinese Academy of Sciences, China.
Recent Advances on Hand Rehabilitation Robots for Post-Stroke Patients
Abstract: Post-stroke patients pay most attention to the upper-/lower-limb rehabilitation and neglect the rehabilitation training of the hand. However, hand is the most important execution organ of human beings, which plays a critical role in daily lives. Meanwhile, the area charging the hand motor in the human’s brain is large. Therefore, the study on the hand rehabilitation robot can help the function recovery of patients’ hands and improve their brain plasticity, which is valuable theoretically and practically. This talk is going to introduce the mechanism design and optimization techniques of the motion-compatible hand rehabilitation robot to ensure the comfortable and safe use of the robot. In addition, some novel impedance control algorithms are presented to realize the passive/active rehabilitation training.
Biography: Long Cheng received the B.S. (Hons.) degree in control engineering from Nankai University, Tianjin, China, in 2004, and the Ph.D. (Hons.) degree in control theory and control engineering from the Institute of Automation, Chinese Academy of Sciences, Beijing, China, in 2009. He is currently a Full Professor with the Institute of Automation, Chinese Academy of Sciences. He is also an adjunct Professor with University of Chinese Academy of Sciences. He has published over 100 technical papers in peer-refereed journals and prestigious conference proceedings. He was a recipient of the IEEE Transactions on Neural Networks Outstanding Paper Award from IEEE Computational Intelligence Society, the Aharon Katzir Young Investigator Award from International Neural Networks Society and the Young Researcher Award from Asian Pacific Neural Networks Society. He is currently serving as an Associate Editor/Editorial Board Member of IEEE Transactions on Cybernetics, Neural Processing Letters, Neurocomputing, International Journal of Systems Science, and Acta Automatica Sinica. His current research interests include the rehabilitation robot, intelligent control and neural networks.