Structured Intelligence for Robot Manipulation

Abstract

Contact-rich manipulation requires compliant bodies for safe, adaptive interaction, yet remains challenging because robots must reason about real-world complexity and uncertainty in real time. Learning-based approaches have shown impressive effectiveness, and scaling laws suggest the power of more data, yet their internal mechanisms remain poorly understood. This talk presents a perspective on building structured intelligence for physical interaction. We begin with compliant and proprioceptive actuation for robotic manipulation, showing how soft-rigid hybrid structures can retain the safety and adaptability of soft materials while gaining the sensing, observability, and control precision needed for dexterous manipulation. We then examine structured tactile sensing and visual-tactile learning as ways to recover useful physical information from complex contact signals. Finally, we turn to robot learning: rather than assuming that more randomized data necessarily produces better policies, we propose the empirical Neural Tangent Kernel analysis to distinguish memorization, adaptation, and invariance to irrelevant variation. Across mechanism design, perception, and learning, we advance structure-conditioned intelligence: robot capabilities are shaped by physical structure in their bodies, their sensory representations, and their learning policies, enabling robust and generalizable manipulation. 

Biography

Chao Liu is an Assistant Professor in the Department of Mechanical Engineering and an Associate Member in the Department of Computer Science at The University of British Columbia, Vancouver. Liu was a Postdoctoral Associate at the Computer Science and Artificial Intelligence Laboratory (CSAIL) at MIT, hosted by Daniela Rus. Before this, Liu completed his Ph.D. advised by Mark Yim at the ModLab, a part of the General Robotics, Automation, Sensing & Perception (GRASP) Laboratory at the University of Pennsylvania. His research focuses on bio-inspired robots, swarm and modular robots, parallel robots, robotic control and motion planning, and grasping and manipulation through multimodal sensing. His work includes the hardware design of several robotic platforms, novel perception solutions, and efficient algorithms on motion planning and control. Liu was the finalist for the Best Paper Award on Safety, Security, and Rescue Robotics in Memory of Motohiro Kisoi at IROS 2019. 

Join us in person, or online for a talk by Prof. Eiichi Yoshida from Tokyo University of Science. 

Link to Talk

Abstract: Recent advent of artificial intelligence and the rapid progress on motion capacity of humanoid robots are bringing a strong attention expecting industrial and social applications. Despite their highly dynamic motion ability, humanoid robots need further improvements in motions involving contact with their body. We humans generate such contact-rich motions with ease while solving the complex problem of combined discrete contact sequence and continuous dynamic motions is extremely hard. In this talk, I will present ongoing research to address the following challenges: collecting whole-body contact motion data from humans, learning behaviors from those data and synthesizing multi-contact humanoid motions. 

Biography: Prof. Eiichi Yoshida is a Professor at the Tokyo University of Science since 2022. He received PhD. degree from the Graduate School of Engineering, the University of Tokyo in 1996. He then joined the National Institute of Advanced Industrial Science and Technology (AIST), Japan, and served as Co-Director of AIST-CNRS Joint Robotics Laboratory from 2004 to 2021, at LAAS-CNRS, Toulouse, France and AIST, Japan. He is a Fellow of IEEE and received awards including Best Paper Award in Advanced Robotics Journal, and the honor of a national medal from the French Government. His research interests include humanoid robotics, human modeling, and task and motion planning for robots. 

Title:
Physical AI and Intelligent Systems: Research Infrastructure and Collaborative Opportunities at the ACIS Laboratory

Abstract:
Artificial Intelligence has rapidly evolved in the digital world, yet its integration into the physical world, for example to robustly execute robots and autonomous systems, demands advanced scientific and technological insights into complex, real-world systems. This presentation highlights the capabilities and collaborative potential of the Advanced Control and Intelligent Systems (ACIS) laboratory at the University of Victoria.

ACIS@UVic operates across three specialized divisions:
•    Robotics and Autonomous Systems: Driving innovations in autonomous mobility, unmanned aerial systems (UAS), and complex field robotics.
•    Physical AI: Bridging the gap between machine learning and physical dynamics, focusing on hardware-in-the-loop validation, human-centric settings such as XR and HRC technologies.
•    Intelligent Systems: Delivering data-driven and AI-powered solutions for diagnosis and prognosis tools, operations research, and complex scheduling problems.

In this talk, Dr. Najjaran will demonstrate the core strengths in training HQP, experimental facilities, and mastering the process of building, curating, and structuring high-fidelity datasets specifically for physical AI applications. By showcasing the lab's current portfolio of AI-driven research, this session aims to identify strategic, high-impact research synergies and foster collaborative partnerships.
 

Bio:
Dr. Homayoun Najjaran is a Professor in the Faculty of Engineering and Computer Science at the University of Victoria, where he serves as the Principal Investigator (PI) of the Advanced Control and Intelligent Systems (ACIS) laboratory. His research and teaching focus on the analysis and design of mechatronics, control systems, and applied artificial intelligence, with broad applications in industrial automation, robotics, and autonomous systems. Together with his students, he has published over 200 articles dedicated to ensuring the safe and robust operation of autonomous systems in industrial settings. Dr. Najjaran is a registered Professional Engineer (P.Eng.), a Fellow of the CSME, and the CEO of Cognia AI, a company providing AI-powered solutions and services to industry.