新闻动态Activité
Activity

学术讲座Towards Safe Robotic Interaction with Unknown Environments

浏览量:时间:2026-09-16

活动主题:Towards Safe Robotic Interaction with Unknown Environments

活动类型:学术交流

举办单位:智能系统与人形机器人国际研究中心

活动时间:2026-9-23 10:00-11:00

活动地点:科研三号楼2111

面向群体:全院师生

主讲嘉宾:

LIANG Wenyu is currently a Senior Scientist and a Team Lead in the Embodied AI and Robotics Chapter at the Institute of Advanced Intelligence and Computing (IAIC), A*STAR, Singapore. He is also an Adjunct Assistant Professor in the Department of Electrical and Computer Engineering, National University of Singapore (NUS). He received his PhD in Electrical and Computer Engineering from the NUS in 2014. He has co-authored over 100 publications, including 2 monographs, 2 book chapters,  and more than 60 journal papers, and holds 2 granted patents. He has received many innovation-related awards, such as the A*STAR CDF Day 2024 Outstanding Presentation under Mechanical, Robotics and Engineering Track, the AAAI 2021 Best Demonstration Award, the KUKA Innovation Award 2021 Finalist, the IES Prestigious Engineering Achievement Awards 2016: Young Creators, etc. His research interests mainly include robotics, intelligent systems, dexterous manipulation, precision motion control, and force control.  

内容摘要:

As robots increasingly operate beyond structured industrial settings and physically interact with humans and complex environments, ensuring both safety and task performance becomes a fundamental challenge. In many real-world applications, the properties of the environment, such as geometry, dynamics, and stiffness, may be uncertain or even completely unknown, while unexpected disturbances and unpredictable movements can further complicate physical interaction. Therefore, to achieve robust task performance while maintaining safety, robots are required to continuously adapt their interaction behavior in response to changing and uncertain contact conditions.

In this talk, a series of our recent research efforts towards safe robotic interaction with unknown environments will be presented. Our work investigates adaptive, robust, and learning-based approaches that enable robots to adapt their interaction behavior to uncertain and changing contact conditions, while balancing safety, adaptability, and task performance. In particular, different interaction control strategies, including variable impedance control, model predictive control, and learning-based control will be introduced in this talk.  

联系人:智能系统与人形机器人国际研究中心,潘婷婷