直播时间：2025年6月3日（周二）20:00-21:30

直播平台：

科学网APP

https://weibo.com/l/wblive/p/show/1022:2321325173485889781903

（科学网微博直播间链接）

科学网微博

科学网视频号

北京时间6月3日晚八点，iCANX Youth Talks第100期邀请到了帝国理工学院生物医学机器人讲师张丹丹，北京大学助理教授韩梦迪，哈尔滨工业大学（深圳）教授邵奕天担任主讲，大连理工大学教授刘倩担任研讨嘉宾，北京航空航天大学教授王党校担任主持人，期待你一起加入这场知识盛宴。

【嘉宾介绍】

张丹丹

帝国理工学院

用于具身智能的多模态传感器设计与制造

【Abstract】

Achieving embodied intelligence in robotic systems requires perceptual capabilities that are human-like, context-aware, and physically grounded. This talk presents our recent advances in the design, fabrication, and integration of multimodal sensors tailored to enable intelligent robots to perceive, reason, and interact dexterously within unstructured, contact-rich environments.We begin with VitacTip, a biomimetic vision-tactile fusion sensor that emulates the integrated sensory capabilities of human skin. By combining vision and tactile sensing within a compact elastomer-based structure, VitacTip enables high-fidelity perception of contact geometry, texture, and pressure. It supports robust, human-like interaction under uncertainty and in dynamically changing environments. Built on this, our multi-modality sensing framework uses generative learning for real-time cross-modal fusion, enhancing robustness and accuracy in sensor interpretation.Next, we introduce the CrystalTac family—vision-based tactile sensors (VBTSs) designed with on-demand sensing mechanisms and fabricated using a rapid monolithic multimaterial 3D printing technique. CrystalTac sensors incorporate embedded 3D fiducial grids within soft elastomers, allowing high-resolution reconstruction of normal and shear forces via internal camera tracking. The platform emphasizes scalability, design flexibility, and ease of integration into robotic systems ranging from macro-scale manipulators to microscale tools.Within the CrystalTac family, we highlight MagicTac, a novel high-resolution VBTS that employs a 3D multi-layer grid structure inspired by a Magic Cube. This structure enhances spatial resolution for external contact perception, while its fabrication—via multi-material additive manufacturing—ensures cost-effectiveness, reproducibility, and minimal assembly requirements. We evaluate MagicTac using deformation field and optical flow-based reconstruction, demonstrating its precision in contact sensing.These sensing technologies are integrated into compliant robotic end-effectors. The TacFR-Gripper, a reconfigurable fin-ray gripper with embedded tactile skin, enables adaptive in-hand manipulation with continuous feedback. Expanding on this, we present MagicGripper, a compact and fully integrated end-effector that combines miniaturized MagicTac, visual perception, and proximity sensing. This multimodal configuration empowers robots to simultaneously perceive contact dynamics, object pose, and motion trajectories in real time.We validate these systems across a range of robotic tasks, including teleoperated assembly, contact-based alignment, and autonomous grasping in cluttered scenes. The results show improved localization accuracy, force estimation, and overall manipulation robustness.Together, these contributions represent significant progress toward the realization of perceptually rich, physically embodied intelligence in robotics. Through innovation in sensor architecture, rapid fabrication, and multimodal integration, we enable the next generation of intelligent robots to interact with the world with human-like dexterity, adaptability, and context awareness.

在机器人系统中实现具身智能，需要具备类人、具有情境感知能力并与物理环境紧密耦合的感知能力。本报告介绍了我们在多模态传感器设计、快速制造及系统集成方面的最新进展，旨在赋予智能机器人在非结构化、接触密集环境中进行感知、推理与灵巧交互的能力。我们首先介绍VitacTip，这是一种仿生视觉-触觉融合传感器，模拟人类皮肤的集成感知能力。VitacTip 通过将视觉与触觉感知集成于紧凑的弹性体结构中，能够高保真地感知接触的几何形状、纹理与压力，适用于动态、不可预知的复杂环境。基于此，我们进一步开发了一个多模态感知框架，结合生成式学习，实现传感器数据的实时跨模态融合与解释，显著增强了系统的感知稳健性与泛化能力。接下来，我们提出CrystalTac传感器家族，这是一类基于视觉的触觉传感器（VBTS），具有按需设计的感知机制，并采用多材料一体化3D快速打印工艺制造。CrystalTac 在柔性弹性体中嵌入三维标记网格，利用内部摄像头进行追踪，从而实现对法向力和切向力的高分辨率重建。该平台具有良好的可扩展性与结构灵活性，适用于宏观机械臂与微尺度操作工具。在此基础上，我们重点介绍MagicTac，这是 CrystalTac 系列中一款采用“魔方”结构灵感设计的高分辨率视觉触觉传感器。MagicTac 采用三维多层网格结构，显著提升了空间分辨率，可精准感知复杂外部接触信息。传感器通过多材料增材制造技术一次成型，具备低成本、可重复制造与低组装门槛等优势。我们通过变形场与光流重建实验对其性能进行了验证，展现出优越的触觉感知能力。这些传感器系统已成功集成至柔顺型机器人末端执行器中。例如，TacFR-Gripper 是一款可重构的鱼鳍状夹爪，内嵌触觉皮肤，支持自适应的手内操控与实时反馈。在此基础上，我们开发了MagicGripper，这是一款高度集成的末端执行器，结合了微型 MagicTac、视觉感知与近距离感应模块，能够实时同时感知接触动态、物体姿态及运动轨迹。我们在多种任务中验证了上述系统的性能，包括远程装配、接触对齐与复杂环境中的自主抓取。结果表明，在定位精度、力感知与操作鲁棒性方面均取得显著提升。综上所述，我们的研究为实现感知丰富、物理具身的智能机器人迈出了关键一步。通过传感器架构创新、快速一体化制造与多模态系统集成，我们推动下一代机器人具备类人般的灵巧性、适应性与情境感知能力，以胜任复杂、人本导向的实际任务场景。

【BIOGRAPHY】

Dr. Dandan Zhang is a Lecturer (US equivalent: Assistant Professor) in Biomedical Robotics with the Department of Bioengineering, and a Lecturer in Artificial Intelligence and Machine Learning with the I-X initiative at Imperial College London. She has cross-disciplinary interests in robotics, biomedicine, and machine learning. Her current research focuses on dexterous manipulation integrated with multi-modality sensor fusion and physical/artificial intelligence, as well as microrobotics for biomedical applications. Dr. Zhang’s academic excellence is reflected in her extensive publication record, with over 40 peer-reviewed papers in leading international journals and conferences, including IEEE Transactions on Robotics, IEEE Robotics and Automation Magazine, Science Robotics (AAAS), Advanced Science, Communications Physics, Matter, ACS Photonics, and Advanced Optical Materials. Her impact is further underscored by the prestigious awards she has received, including the UK Best PhD in Robotics Thesis Award for her pioneering doctoral research and the Amazon PhD Prize for Outstanding Achievements in Robotics. As a first author, she has also been honored with best paper awards at top-tier international conferences, such as the IEEE International Conference on Robotics and Automation (ICRA), and has received finalist distinctions at IEEE IROS and MARSS, highlighting the depth and quality of her contributions to robotics and AI. In 2025, Dr. Zhang was selected as a representative for the AI Visionaries: Leading Health Innovation for Women initiative in UK. This initiative recognizes leading female researchers whose work is shaping the future of AI in healthcare and biomedical engineering.

张丹丹博士是伦敦帝国理工学院生物工程系的生物医学机器人讲师，同时也是该校 I-X 人工智能与机器学习项目的讲师。她的研究跨越机器人、生物医学与机器学习等多个学科领域。目前，她的研究重点是将多模态传感融合与物理/人工智能相结合，以实现灵巧操作，并推进面向生物医学应用的微型机器人技术。她的目标是提升多尺度机器人系统的自主性水平，推动下一代精准医疗与个性化家庭护理的发展。张博士在学术研究方面表现卓越，已在国际顶尖期刊与会议上发表超过 40 篇同行评审论文，涵盖《IEEE Robotics and Automation Magazine》、《IEEE Transactions on Robotics》、《Science Robotics》（AAAS 出版）、《Advanced Science》、《Communications Physics》、《Matter》、《ACS Photonics》和《Advanced Optical Materials》等。她的学术影响力还体现在多项重要奖项的获得上，包括因其开创性的博士研究而荣获的“英国最佳机器人学博士论文奖”，以及“亚马逊杰出机器人研究博士奖”。作为第一作者，她的研究还获得了包括 IEEE 国际机器人与自动化会议（ICRA）在内的多个顶级国际会议的最佳论文奖，并在 IEEE IROS 和 MARSS 等会议上获得了优秀论文入围奖，彰显了她在机器人与人工智能领域的研究深度与广度。2025 年，张博士入选英国“AI Visionaries: Leading Health Innovation for Women”女性健康创新领袖计划，旨在表彰在人工智能与生物医学工程领域推动健康创新的杰出女性研究者。

韩梦迪

北京大学

基于三维微型应变片的电子皮肤

【ABSTRACT】

Flexible tactile sensors play important roles in many areas, like human-machine interface, robotic manipulation, and biomedicine. However, their flexible form factor poses challenges in their integration with wafer-based devices, commercial chips, or circuit boards. Here, we introduce manufacturing approaches, device designs, integration strategies, and biomedical applications of a set of flexible, modular tactile sensors, which overcome the above challenges and achieve cooperation with commercial electronics. The sensors exploit lithographically defined thin wires of metal or alloy as the sensing elements. Arranging these elements across three-dimensional space enables accurate, hysteresis-free, and decoupled measurements of temperature, normal force, and shear force. Assembly of such sensors on flexible printed circuit boards together with commercial electronics forms various flexible electronic systems with capabilities in wireless measurements at the skin interface, continuous monitoring of biomechanical signals, and spatial mapping of tactile information. The flexible, modular tactile sensors expand the portfolio of functional components in both microelectronics and macroelectronics.

柔性触觉传感器在人机交互、机器人操控和生物医学等领域具有重要作用。然而其柔性特质使其与晶圆器件、商用芯片或电路板的集成面临挑战。本研究提出了一套柔性模块化触觉传感器的制造方法、器件设计、集成策略及生物医学应用方案，成功实现了与商用电子器件的协同工作。该传感器采用光刻定义的金属/合金细丝作为传感元件，通过三维空间排布这些元件，实现了温度、法向力与剪切力的高精度、无迟滞且解耦的测量。将此类传感器与商用电子器件共同集成于柔性印刷电路板，可构建多种柔性电子系统，实现皮肤界面的无线测量、生物力学信号的连续监测以及触觉信息的空间映射。这种柔性模块化触觉传感器同时拓展了微电子学和宏电子学领域的功能元件体系。

【BIOGRAPHY】

Dr. Mengdi Han is an Assistant Professor in the Department of Biomedical Engineering, College of Future Technology, Peking University. He received his B.S. degree in Huazhong University of Science and Technology in 2012 and Ph.D. degree in Peking University in 2017. He was a visiting Ph.D. student at Department of Materials Science and Engineering, University of Illinois Urbana-Champaign from 2015 to 2017. He worked as a postdoctoral fellow at Querrey Simpson Institute for Bioelectronics, Northwestern University from 2017 to 2020. He published more than 100 SCI-indexed papers, including first/corresponding author papers in Nature Electronics, Nature Biomedical Engineering, Science Translational Medicine, Science Robotics, Science Advances, PNAS, Advanced Materials, etc. His research group aims to develop advanced micromechanical bioelectronics for electronic skins, wireless biosensors and microrobotics. His research has been recognized with many awards including Microsystems & Nanoengineering Young Scientist Award (2020), MIT Technology Review Innovators Under 35 Asia Pacific (2021), World’s Top 2% Scientists by Stanford & Elsevier (2023), iCANX Young Scientist Award (2024), and Asian Young Scientist Fellowship (2024).

韩梦迪，北京大学未来技术学院生物医学工程系助理教授，研究员，博士生导师。2012年本科毕业于华中科技大学电子科学与技术系，2017年博士毕业于北京大学微纳电子学系；2015-2017年作为联合培养博士生就读于美国伊利诺伊大学材料科学与工程系，2017-2020年于美国西北大学Querrey Simpson生物电子研究所从事博士后研究。已发表SCI论文百余篇，其中第一/通讯作者论文发表于Nature Electronics、Nature Biomedical Engineering、Science Translational Medicine、Science Robotics、Science Advances、PNAS、Advanced Materials等期刊。获Microsystems & Nanoengineering 青年科学家（2020）、《麻省理工科技评论》亚太地区35岁以下科技创新35人（2021）、世界顶尖2%科学家（2023）、iCANX青年科学家（2024）、亚洲青年科学家（2024）等奖励及荣誉。韩梦迪课题组致力于研发微机械生物电子器件，助力电子皮肤、无线生物传感、微型机器人等领域的发展。

邵奕天

哈尔滨工业大学（深圳）

宽频触觉信息：人类交互与具身智能感知

【ABSTRACT】

Touch is among the most important senses of humans to perceive and interact with their environment. Even simple touch interactions generate complex tactile signals in the tissues of the human body. Among these, broadband tactile signals propagate over long distances on the body surface in the form of mechanical waves, exhibiting a wide range and heterogeneous spatiotemporal distribution. This talk presents research investigating the dynamic tactile sensing of humans, including how such sensing mechanism permits an efficient encoding of touch information that minimizes the corresponding energy consumption, and how it can inspire novel designs of tactile sensing systems of wearable machines and robots. It introduces the physical principles that generate these tactile signals and the corresponding human tactile mechanisms, together with how to capture and reconstruct the signal distribution in real-time using a distributed wearable tactile sensor array, and how to decode rich tactile interaction information from these signals. It will also introduce tactile sensing systems inspired by human touch sensing mechanisms, explaining how it can imitate human-like multimodal tactile sensing together with its potential applications, including the usage in capturing tactile signals from robot touch and utilizing the signals to perceive the environment or for providing haptic feedback to human users.

触觉感知赋予了人类与环境灵巧交互的能力。然而，即便是简单的触碰交互，也会在人类肢体中产生复杂的触觉信号。其中，宽频的动态触觉信号以机械波的形式在肢体表层远距离传播，具有大范围、异构的时空分布。本报告将分为两个部分。第一部分介绍产生这些动态触觉信号的物理原理和对应的人体触感机理，并展示如何捕捉并还原信号在肢体上分布，以及如何从这些信号中解码出丰富的触觉交互信息。第二部分将介绍受人类动态触感机理启发而设计的智能触觉传感系统，如何使其具备类人的灵敏触觉感知能力，以及在生产和生活中对应的潜在应用。

【BIOGRAPHY】

Yitian Shao directs the HUMIT Lab at the School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen. He is also affiliated with the International Research Institute for Artificial Intelligence and the Cyber-Physical Systems Research Center at Harbin Institute of Technology, Shenzhen. He has long been engaged in research on haptic interfaces, robotic tactile sensing, wearable technologies, and extended reality. From 2022 to 2023, he worked as a junior professor (W1) in the Department of Electrical and Computer Engineering at TU Dresden in Germany. He was also associated with the Cluster of Excellence, Center for Tactile Internet with Human-in-the-Loop (CeTI) and 6G-life research hub in Germany. He was a Postdoctoral Researcher at the Max Planck Institute for Intelligent Systems from 2021 to 2022, and was awarded with Humboldt Research Fellowship. In 2020, he received his Ph.D. degree from the Department of Electrical and Computer Engineering at the University of California, Santa Barbara, US, and won the EHS Best PhD Thesis Award. In 2019, he worked as an intern researcher at Microsoft Research Institute in the US. In 2013, he received his Bachelors degree in Electrical Engineering and Automation from Tianjin University.

邵奕天，博士，洪堡学者，哈尔滨工业大学（深圳）计算机科学与技术学院教授/博导，深圳市鹏城孔雀计划B档特聘岗位。兼任于哈尔滨工业大学（深圳）国际人工智能研究院，信息物理系统CPS研究中心，以及智慧农场技术与系统全国重点实验室。回国前任职于德国德累斯顿工业大学电子与计算机工程学部W1助理教授，兼任德国精英集群中心Centre for Tactile Internet with Human-in-the-Loop (CeTI) 及6G-life研究中心助理教授。以触觉学（Haptics）为基础长期从事机器人触感，人机交互，可穿戴设备等相关方向的研究，成果发表在包括Science Advances，PNAS，Nature Materials, Device (Cell Press) 等知名期刊上。25年发表的Adv. Mater. Technol. 获选封面文章。22年在Springer出版学术专著1部。21年获触觉学领域最佳博士论文奖。在触觉学领域国际顶级会议IEEE World Haptics和IEEE HAPTICS获7个奖项与提名。与美国Meta Reality Labs合作的研究发表了3项专利。20年于美国加州大学圣芭芭拉分校电子与计算机工程系获得博士学位。19年在美国微软研究院任实习研究员。13年于天津大学电气工程及其自动化学院获得工学学士学位。任Frontiers in Robotics and AI评审编委。

【主持人】

王党校

北京航空航天大学

【研讨嘉宾】

刘倩

大连理工大学