Plenary Speakers

Koushil Sreenath

UC Berkeley

October 28(Wed) 10:40-11:40, Palace Hall, 3F

The Bitter Lesson in Control of Legged Robots

The history of legged robot control can be viewed as a gradual shift in how we encode our understanding of locomotion: in models, in controllers, or in data. Early work simplified locomotion through reduced-order models such as the inverted pendulum and control strategies such as zero-moment-point-based control, sacrificing fidelity for tractability. As computation caught up, we could bring far richer, full-body models into the controller, enabling orbital stability of hybrid periodic orbits underlying legged locomotion.

In the past few years, data and computation have increasingly displaced hand-designed structure in our models and controllers.
In this talk, I trace that evolution through my own work. We began in the era of model-based control, where hybrid zero dynamics, control Lyapunov functions (CLFs), and control barrier functions (CBFs) provided explicit guarantees of stability and safety. We then augmented model-based control with learning, pairing these formal tools with reinforcement learning and Gaussian-process-based CLFs and CBFs to let data fill in what our models could not capture. The next step was toward purely learning-based methods, first reinforcement learning and then generative approaches, that trade formal guarantees for greater generalization and robustness. Running through all of this is the bitter lesson: that data and scale, rather than clever control algorithms, tend to win.

Biography

Koushil Sreenath is a Professor and Roscoe and Elizabeth Hughes Chair in Mechanical Engineering at University of California, Berkeley. He received his Ph.D. in Electrical Engineering and Computer Science and a M.S. degree in Applied Mathematics from the University of Michigan at Ann Arbor, MI, in 2011. He was a Postdoctoral Scholar at the GRASP Lab at University of Pennsylvania (2011-2013) and an Assistant Professor at Carnegie Mellon University (2013-2017). His research interest lies in the application of nonlinear control and learning to highly dynamic robotic systems. He received the conference Best Paper Awards at Learning for Dynamics & Control (L4DC), Robotics: Science and Systems (RSS), and IEEE Conference on Robotics, Automation, and Mechatronics (RAM). He also received the NSF CAREER Award, the NSF CRII Award, the Hellman Fellow, and Google Faculty Research Award in Robotics.

Hiroyuki Shinoda

University of Tokyo

October 29(Thu) 10:40-11:40, Palace Hall, 3F

Beyond Touch: High-Fidelity Noncontact Haptics Supporting the Mind

Artificial Intelligence, which is currently revolutionizing workflow automation, will eventually shift its focus toward human beings themselves. Technological priorities are expected to transition toward enhancing human well-being and improving both physical and mental health. In this context, haptics is expected to become an increasingly vital channel for human-computer interfaces; however, haptic technology remains underutilized in IT today.

This talk will discuss the necessity of this shift and the emerging technologies that address it. I will introduce technologies that are rapidly becoming a reality, capable of reproducing and generating haptic sensations with high-fidelity and versatility without requiring any specialized equipment on the user’s body. I will demonstrate how these technologies possess the potential not only to increase operational efficiency but also to support mental health, fostering a healthier and more fulfilling life.

Biography

Hiroyuki Shinoda is Professor and Vice Dean at the Graduate School of Frontier Sciences, the University of Tokyo. He also serves as Vice President of SICE. After receiving his Ph.D., he began his career as an Associate Professor at the Tokyo University of Agriculture and Technology (1995–1999), during which he also spent a year as a Visiting Scholar at the University of California, Berkeley. He was an Associate Professor at the University of Tokyo from 2000 to 2012 and has held his current professorship since 2013.
His group demonstrated the world’s first noncontact ultrasound tactile display in 2008. His primary research focus lies in enhancing the realism and facilitating the social implementation of this technology. His research group has received numerous awards, including the Best Conference Paper Award at IEEE ICRA 1999 and the 2022 IEEE Transactions on Haptics Best Paper Award. He has also held significant leadership roles in the research community, serving as General Chair for Asia Haptics 2016, Program Co-Chair for EuroHaptics 2018, General Co-Chair for the IEEE World Haptics Conference 2019, and Steering Committee Chair for the IEEE World Haptics Conference 2021.

Jean Oh

Carnegie Mellon University

October 30(Fri) 10:40-11:40, Palace Hall, 3F

Creative Physical AI

Do robots need creativity? I will share my stance that they do need creativity to solve general problems and support human values. Physical AI is a type of AI that enables robots to perceive and interact with a physical world. Trendy approaches in physical AI such as Vision-Language-Action (VLA) models directly map the observations to actions where robots make decisions dominantly based on sensed information. While sensing is crucial for understanding the current physical environments, this paradigm of physical AI is fundamentally limited to support general tasks where humans see around corners and solve problems creatively based on not only what they can observe now but also various predictions of the latent spatiotemporal and social contexts. I will illustrate the examples where robots without creativity can fail to fulfill even simple goals and how we can develop physical AI for creative problem solving.

If equipped with creative physical AI, can such robots promote human creativity as in creating arts? Generative AI has brought us numerous types of convenience in the digital art world. To create artifacts in the real world, creative physical AI is needed, for instance, to preserve traditional craftsmanship such as wood carving or claymation, which faces declining participation due to its labor-intensive nature. More broadly, our innovations in creative physical AI aim to encourage people to participate in more creative activities such as educational and therapeutic art sessions. I would like to invite the audience to think about how we can use technologies to promote human creativity for the next generation.

Biography

Jean H. Oh is an Associate Research Professor and Director of roBot Intelligence Group (BIG) at the Robotics Institute at Carnegie Mellon University (CMU). Jean’s research goal is to create robots that can co-exist and collaborate with humans in shared environments, continuously learning to improve themselves over time through training, exploration, and interaction. The philosophical goal of Jean’s research is to develop technologies to remind us of “what makes us human,” promoting humanity such as safety, creativity, and compassion. Jean is best known for her contributions to socially compliant navigation and interdisciplinary physical AI research between robotics and arts. Jean’s work has won numerous best paper awards at various robotics conferences such as IEEE International Conference on Robotics and Automation (ICRA) and ACM/IEEE International Conference on Human-Robot Interaction (HRI), and featured in media worldwide including New York Times, CNET, Arirang TV, ABC, CBS, and KBS. Jean received a PhD from CMU, MS from Columbia University, and BS from Yonsei University. Jean is the founder of Lavoro AI where she brings her creative, safe, and compassionate Physical AI innovations to market.