南开大学人工智能与机器人国际学术讲坛(第一期)


南开大学人工智能与机器人国际学术讲坛

 Nankai University International E-Forum on Artificial Intelligence and Robotics

 第一期  


日期 ——会议ID

时间

主讲人

单位

2020117日(周六)

腾讯会议:356 2733 3629

19:30-20:30

I-Ming Chen (陈义明)

新加坡南洋理工大学

20:30-21:30

Xiang   Chen (陈  翔)

加拿大温莎大学

2020118日(周日)

腾讯会议:866 2460 3481

19:30-20:30

Max QH   Meng(孟庆虎)

南方科技大学

20:30-21:30

Jianwei   Zhang(张建伟)

德国汉堡大学





报告1

报告时间:2020117日(周六)19:30-20:30

腾讯会议:356 2733 3629

报告嘉宾:I-Ming Chen 新加坡南洋理工大学

报告题目:Robotics & Automation for Real World Challenges: From Construction to Kitchen Food Handling

AbstractRobots have used in manufacturing industry for a long time for mass production. With the advancement of 3D vision, machine learning algorithms, and lower hardware cost, robots have started been seen in other challenging scenarios like hospitals, environment cleaning, construction and household. This seminar would introduce some of our recent robotics projects coming from real world demands: construction automation and kitchen food handling. A new R&D Program on Robotics for Construction 4.0 sponsored by Singapore’s National Robotics Program will be introduced first. This program adopts a systematic approach based on Design for Manufacturing and Assembly (DfMA) paradigm for construction robotics and automation, i.e., Construction 4.0, to address the construction productivity issues. The program consists of four major works using AI, Robotics to address the productivity in pre-cast factories as well as onsite installation and other construction tasks. Second, we present a systematic approach to handle the preparation and assembly of in-flight food in the catering center using 3D vision, machine learning and hybrid/soft grippers and the robot. The system approach used would be a model for other types of real-world automation projects.

报告人简介:

陈义明教授,台湾大学本科毕业,美国加州理工学院硕士与博士。现为南洋理工大学终身正教授。陈教授长期从事可重构机器人系统以及人机交互研究工作,是可重构机器人系统方面的先驱。他在这个领域中的工作为大规模可重构机器人与自动化系统的实现与关键驱动模块技术的发展奠定了基础,对国际机器人学学术界与工业界产生了重要的影响,并享有很高的国际声誉。他近期研究以物流机器人与建筑机器人应用研究为主。曾任南洋理工大学机器人中心主任与智能系统中心主任,新加坡国家研究基金机器人产业规划工作小组成员,新加坡科技研发署工业机器人项目副负责人,台湾工业技术研究院智能机器人计划顾问,一些国际知名机器人企业技术顾问与并购顾问,多所大学的讲席教授,国家自然科学基金委员会双清论坛海外专家。目前为新加坡工程院院士,美国机械工程学会会士(ASME Fellow),与美国电气和电子工程学会会士(IEEE Fellow)IEEE/ASME Transactions on Mechatronics期刊总主编,新加坡Transforma Robotics Pte Ltd(建筑机器人)的创始人与首席执行官, Hand Plus Robotics Pte Ltd (物流机器人)的首席技术官。


报告2

报告时间:2020117日(周六)20:30-21:30

腾讯会议:356 2733 3629

报告嘉宾:Xiang Chen 加拿大温莎大学

报告题目:Multi-objective H2 and H Filtering and Control--A New Paradigm

Abstract A new design paradigm is discussed in this talk which allows multi-objective filtering and control designs to achieve complement H2 and H∞ performance with little trade-off. In particular, a revisit of Youla-Kucera parameterization of all stabilizing controllers and the traditional mixed H2/H∞ filtering and control are first presented. Then the new design paradigm is introduced in comparison with the traditional structure. It is shown that the new paradigm is not only able to automatically render the H2 control performance if there is no modeling mismatch for the plant, but also provide recovery, instead of compromise, of the optimal performance when the modeling error is present, noting that the compromise is normally seen in traditional mixed designs. It is also noted that the recovery of the robust performance is regulated by the ‘measured error size’ of the modeling mismatch, hence, resulting in less conservativeness of the control performance. An inverted pendulum example is presented to validate the design expectations of the new control paradigm.

报告人简介:

陈翔教授,1998年在路易斯安那州立大学获得博士学位。现为加拿大温莎大学(University of Windsor)电子与计算机工程系教授。顶级期刊《IEEE/ASME Transaction on Mechatronics》高级编辑,《SIAM Journal on Control and Optimization》,《International Journal of Intelligent Robotics and Applications》,《Control Theory and Technology》和《Unman Systems》的副编辑。他系统地建立了非线性(静态和动态)系统的镇定与鲁棒控制方法,证明解决了此类系统的鲁棒镇定控制这一公认的难题,在多目标鲁棒控制理论方面做出了奠基性的重要贡献;在视觉传感系统的优化方面开展了一系列基础性与应用型的开创性研究,取得了世界领先的成果。陈翔教授4次获得温莎大学研究奖,共发表100多篇论文。他的主要研究方向包括:具有复杂性系统的鲁棒与最优控制、复杂网络化系统、汽车控制系统、机器视觉传感器网络。


报告3

报告时间:2020118日(周日)19:30-20:30

腾讯会议:866 2460 3481

报告嘉宾:Max QH Meng  南方科技大学

报告题目:机器人可以达到人类的智慧吗?

Abstract 人工智能正在成为人类生活不可分割的一个重要部分。本讲座结合南科大孟庆虎团队在场景智能和机器人领域勤耕深挖逾三十载的沉淀积累,报告在智慧医疗、手术机器人、以及稠密人群动态环境中交互操作服务机器人等领域的研究和产业化成果。最后对人工智能和机器人的未来发展提出个人拙见以抛砖引玉。

报告人简介:

孟庆虎,加拿大工程院院士,Fellow of IEEE,中组部国家特聘专家,广东省高尖端人才,深圳市杰出人才。1992 年获加拿大维多利亚大学博士学位。1994-2001 年在加拿大阿尔伯塔大学任教至终身正教授。2001 年起任香港中文大学教授,2016年起任电子工程学系主任。2020年起任南方科技大学电子与电气工程系讲席教授、系主任。研究领域涉及机器人感知与智能及智慧医疗等。多个研究课题在国际上独创建树,国际领先。担任多个国际学术期刊主编和编委,以及国际会议大会主席,包括国际机器人与自动化旗舰会议 IEEE ICRA 2021 大会主席。


报告4

报告时间:2020118日(周日)20:30-21:30

腾讯会议:866 2460 3481

报告嘉宾:Jianwei Zhang 德国汉堡大学

报告题目:Crossmodal Learning Approaches to Robust Autonomous Robot Systems

AbstractIn the past covid-19 era, AI and robotics need to be combined to solve some real-world challenges by combining machine automation with realization of cognitive abilities in IT systems. There has been substantial progress in deep neural networks and AI in terms of individual data-driven benchmarking and passing the Turing test, for instance for language translation, face and speech recognition, writing poems or reports, etc. However, such existing data-driven systems are not yet crossmodal, they are not robust in a dynamic and changing world, they do not learn continuously and they do not possess symbolic explainability. For many AI-related tasks, such as multimodal communication, autonomous driving or human-robot collaboration, robust continual learning, processing of multisensory information and the explainability of the learned knowledge and behavior is essential.

My talk will first I introduce a model which allows a robot to better understand new situations by integration of knowledge, planning and learning and then the necessary modules to enhance the robot intelligence level. Then I will demonstrate how a robot can evolve its model as a result of learning from experiences; and how such cross-modal learning methods can be realized in intelligent robots. To evaluate success for a given task in operational service robot platforms with grasping facilities in a restaurant service scenario, we measure the compliance of the actual robot behavior to the intended ideal behavior for that task in that scenario by a measure of ‘‘Distance to Ideal Model’’. Additionally, we also measure the Description Length of the instructions given to the robot to achieve a goal matters. Our general aim of designing learning and reasoning tools for a robot to autonomously and effectively increase its competence was operationalized as: make it possible for a robot to collect experiences allowing it to perform at lower Distance to Ideal Model and shorter Description Length. Then I will introduce several novel intelligent robot systems with potential applications in home and logistic services as well as in manufacturing and bio-medical engineering. At the end, I will summarize the scientific challenges to make a robot behave robustly as an autonomous system.

报告人简介:

Jianwei Zhang is professor and director of TAMS, Department of Informatics, Universität Hamburg, Germany. He also served as member of Chair-Professors in Human-Computer Interaction and Cognitive Neuroscience in the Department of Computer Science of Tsinghua University. He received both his Bachelor of Engineering (1986, with Distinction) and Master of Engineering (1989) at the Department of Computer Science of Tsinghua University, Beijing, China, his PhD (1994) at the Institute of Real-Time Computer Systems and Robotics, Department of Computer Science, University of Karlsruhe, Germany, and Habilitation (2000) at the Faculty of Technology, University of Bielefeld, Germany. His research interests are sensor fusion, intelligent robotics and multimodal machine learning, cognitive computing of Industry4.0, etc. In these areas he has published about 300 journal and conference papers, technical reports, six book chapters and three research monographs. He holds 40+ patents on intelligent components and systems. He is the coordinator of the DFG/NSFC Transregional Collaborative Research Centre SFB/TRR169 “Crossmodal Learning”, the International Research Training Group CINACS (both with Tsinghua University) and several EU and German robotics projects. He has received multiple best paper awards. He is the General Chairs of IEEE MFI 2012, IEEE/RSJ IROS 2015, and the International Symposium of Human-Centered Robotics and Systems 2018. Jianwei Zhang is life-long Academician of Academy of Sciences in Hamburg.