IEEE/RSJ International Conference

on Intelligent Robots and Systems

19 – 25 October, 2025

Hangzhou, CHINA

Plenary Sessions

Plenary Sessions


Marco Hutter

Song-Chun Zhu

  Chair Professor, Peking University, China & Tsinghua University, China

  Founding Director, Beijing Institute for General Artificial Intelligence

TongBrain: Bridging Physical Robots and AGI Agents

Time:   Oct 21, 9:00 – 10:00 AM
Venue:   Exhibition Hall 4D
In this talk, I will present recent research on Robotics and Artificial General Intelligence (AGI) conducted at Beijing Institute of General Artificial Intelligence (BIGAI). I will begin by discussing our work on visual scene understanding, cognitive reasoning, and task and motion planning, which together enable robots to engage in complex physical and social interactions. These developments build upon a progressively deepening cognitive architecture and value system that mark a paradigm shift in the robotics and computer vision communities—from modeling pixels (image-centric) to modeling scenes (object-centric), and to modeling cognitive aspects (agent-centric). Then, I will introduce ‘TongTong,’ a digital AGI agent that extends this paradigm into a unified agentic framework defined by three interconnected components: the cognitive architecture (C), potential functions (U) representing skills, and value functions (V). Various AGI systems can be characterized as points within this joint (C, U, V) space. A new benchmarking framework and testing platform, TongTest, is also presented for measuring the general intelligence of various AI agents on performing multi-modal embodied tasks in complex environments. It assesses the intelligence of TongTong to match a 5-6 years old child. Finally, I will showcase TongBrain which encapsulates a series of Real2Sim2Real efforts: digitizing tens of thousands of real scenes to train TongTong and transferring the learned skills to real-world robotic systems under various application scenarios. These advances provide a foundation for exploring higher-level topics such as AGI/robot morality, social norms, and safety, ultimately paving the way toward a symbiotic society of interconnected humans, physical robots and digital agents.

Speaker Biography

Song-Chun Zhu is currently director of Beijing Institute for General Artificial Intelligence (BIGAI) – an independent non-profit research organization, and Chair Professor at Peking University and Tsinghua University jointly. He is dean of both School of AI and interdisciplinary research Institute for AI at Peking University.    He received Ph.D. degree from Harvard University in 1996, and after that he worked at Brown, Stanford and UCLA where he established an inter-disciplinary center of Vision, Cognition, Learning and Autonomy (VCLA@UCLA 2002-2020) before he returned to China in late 2020.  He has published 400+ articles in AI areas including CV, Cognition, Robotics, ML, NLP, MAS et al.  His scientific contributions have garnered various recognitions, including the Marr Prize (2003),  twice Marr prize honorary nominations (1999, 2007), Sloan Fellowship (2001), J.K. Aggarwal Prize (2008), Helmholtz Test-of-Time Award (2013) , computational modeling prize at CogSci (2017) etc.  He established the Lotus Hill Institute in 2005 to launch large-scale image annotation and pioneered data-driven statistical approaches. He served as general chair for CVPR 2012 and 2019, and led twice the Multi-University Research Initiatives (MURI 2010-2015, 2015-2020) on scene understanding, visual commonsense reasoning and robot autonomy in the US. His research has been motivated by the pursuit of a general unified theory of intelligence.

Marco Hutter

Marco Hutter

  Professor, ETH Zurich, Switzerland

  Director of Center for Robotics, ETH Zurich

The New Era of Mobility: Humanoids and Quadrupeds Enter the Real World

Time:   Oct 22, 9:00 – 10:00 AM
Venue:   Exhibition Hall 4D
Legged robotics has made remarkable strides in recent years, with quadrupeds and humanoids beginning to demonstrate real value in practical applications. Beyond the rapid advances in hardware, breakthroughs in perception, navigation, planning, and reinforcement learning for locomotion have unlocked unprecedented levels of mobility and autonomy on challenging terrain. In this talk, I will explore how reinforcement learning and autonomy are transforming the capabilities of legged robots and other mobile machines. I will share insights into the underlying methodologies, highlight real-world deployments "in the wild," and discuss where we stand today in the long-envisioned journey toward a ubiquitous robotic workforce.

Speaker Biography

Marco is a professor for robotic systems and director of the Center for Robotics at ETH Zurich. Since 2024, he leads the Zurich office of the RAI Institute. His research interests are in the development of novel machines and machines and their intelligence to operate in rough and challenging environments. Together with his team, he realized a number of legged robots, mobile manipulators, and autonomous excavators that find applications including industrial inspection, construction and forest operations, household assistance, and extraterrestrial exploration. Marco is a co-founder of several ETH Startups that commercialize technology ranging from legged robots and autonomous construction equipment.

Hyoun JIN Kim

Hyoun JIN Kim

  Professor, Seoul National University (SNU), Korea

  Director of Automation and Systems Research Institute, Seoul National University

Autonomous Aerial Manipulation: Toward Physically Intelligent Robots in Flight

Time:   Oct 23, 9:00 – 10:00 AM
Venue:   Exhibition Hall 4D
"Aerial robotics is rapidly evolving from platforms that merely fly to systems capable of physical interaction, manipulation, and collaboration. The emerging field of aerial manipulation marks a paradigm shift—enabling flying robots not only to navigate through space but to act upon and reshape it. This talk will explore the expanding frontier of physically intelligent aerial robotics, bridging dynamics, control, perception, and mechanical design to realize truly autonomous aerial interaction. At the heart of this transformation lies the ambition to achieve full autonomy in physical interaction—to manipulate and exert forces at arbitrary poses with precision and robustness. From innovative aerial platforms to fully dexterous multi-DOF manipulators, new control architectures are addressing the coupled dynamics of flight and contact. Advances in time-varying force control, hybrid control for improved transient performance, and robust trajectory tracking have begun to demonstrate stable, force-aware interactions. Beyond individual controllers, the field is exploring whole-body planning and control, where the aerial vehicle and manipulator act as a single, dynamically consistent system. This paradigm enables contact-rich behaviors—including inspection, maintenance, and cooperative transport—while preserving flight stability and efficiency. At a higher level, aerial robotics is becoming increasingly adaptive through learning-based skill generalization, allowing robots to acquire and transfer skills across diverse settings. Ultimately, aerial manipulation represents a new stage in the evolution of flying robots—merging flight and contact, autonomy and intelligence, and transforming them from passive observers into active, intelligent systems capable of shaping their environment."

Speaker Biography

Hyoun JIN Kim is a Professor in the Department of Aerospace Engineering and Director of the Automation and Systems Research Institute at Seoul National University (SNU), Korea, where she is also affiliated with the Graduate School of Artificial Intelligence. She leads the Laboratory on Autonomous Robotics Research, focusing on motion planning, control, and learning-based approaches for aerial and field robotic systems. She received her B.S. from the Korea Advanced Institute of Science and Technology (KAIST) and her M.S. and Ph.D. from the University of California, Berkeley, all in Mechanical Engineering. Before joining SNU in 2004, she was a Postdoctoral Researcher and Lecturer in Electrical Engineering and Computer Sciences at UC Berkeley. At SNU, she has served as Director of the Intelligent Unmanned Mobility Research Center, Chair of the Department of Aerospace Engineering, and Director of the Global Education Center. She has also contributed to governmental and industrial advisory committees, including the Korean Ministry of Strategy and Finance, the Ministry of Land, Infrastructure and Transport, Samsung Research, Hyundai Transys, and Hanwha Aerospace. In the robotics community, Professor Kim serves as an Editor for Robot Learning at the IEEE International Conference on Robotics and Automation (ICRA) and has been an Associate Editor for the IEEE Transactions on Robotics (T-RO), The International Journal of Robotics Research (IJRR), and the Journal of Intelligent & Robotic Systems (JINT). She has also co-chaired the IEEE Technical Committee on Aerial Robotics and Unmanned Aerial Vehicles. She is a member of the National Academy of Engineering of Korea (NAEK) and serves on the National Artificial Intelligence Strategy Council of Korea.