Innovation in (Soft) Robotics and Control
ETH Distinguished Seminar in Robotics, Systems, and Control (151-0623-00)
Welcome to the (soft) Robotics, Vision, and Controls Talks series hosted by ETH Zürich. These open talks focus on innovations in (soft) robotics, computer vision, and control systems, and are held in a hybrid format at ETH Zürich’s main campus.
About the Series
Schedule:
- Fridays at 16:00 CET in HG G5 (in-person) or via Zoom
- Location: ETH Zürich
- Join via Zoom
Next Talk
Developing Generalist Robots
Speaker: Prof. Dr. Chelsea Finn
Affiliation: Stanford University, USA
Date: October 31, 2025
Time & Location: 16:00 ; Zoom — (optional HG G5)
Abstract
While we have seen exciting advances in developing foundation models for text and image modalities, developing useful general-purpose intelligence for robots presents unique challenges. In this talk, I’ll discuss some of our research on data, architectures, and algorithms for developing and evaluating robotic foundation models. This includes enabling robots to complete tasks in completely novel homes through π-0.5, augmenting such systems to have long-term memory, and using generative world models to evaluate and improve such models.
Bio
Chelsea Finn is an Assistant Professor in Computer Science and Electrical Engineering at Stanford University, the William George and Ida Mary Hoover Faculty Fellow, and a co-founder of Physical Intelligence (Pi). Her research interests lie in the capability of robots and other agents to develop broadly intelligent behavior through learning and interaction. To this end, her work has pioneered end-to-end deep learning methods for vision-based robotic manipulation, meta-learning algorithms for few-shot learning, and approaches for scaling robot learning to broad datasets. Her research has been recognized by awards such as the Sloan Fellowship, the IEEE RAS Early Academic Career Award, and the ACM doctoral dissertation award, and has been covered by various media outlets including the New York Times, Wired, and Bloomberg. Prior to joining Stanford, she received her Bachelor’s degree in Electrical Engineering and Computer Science at MIT and her PhD in Computer Science at UC Berkeley.
All Upcoming Talks
Developing Generalist Robots
Speaker: Prof. Dr. Chelsea Finn
Affiliation: Stanford University, USA
Date: October 31, 2025
Time & Location: 16:00 ; Zoom — (optional HG G5)
More info
Abstract
While we have seen exciting advances in developing foundation models for text and image modalities, developing useful general-purpose intelligence for robots presents unique challenges. In this talk, I'll discuss some of our research on data, architectures, and algorithms for developing and evaluating robotic foundation models. This includes enabling robots to complete tasks in completely novel homes through π-0.5, augmenting such systems to have long-term memory, and using generative world models to evaluate and improve such models.
Bio
Chelsea Finn is an Assistant Professor in Computer Science and Electrical Engineering at Stanford University, the William George and Ida Mary Hoover Faculty Fellow, and a co-founder of Physical Intelligence (Pi). Her research interests lie in the capability of robots and other agents to develop broadly intelligent behavior through learning and interaction. To this end, her work has pioneered end-to-end deep learning methods for vision-based robotic manipulation, meta-learning algorithms for few-shot learning, and approaches for scaling robot learning to broad datasets. Her research has been recognized by awards such as the Sloan Fellowship, the IEEE RAS Early Academic Career Award, and the ACM doctoral dissertation award, and has been covered by various media outlets including the New York Times, Wired, and Bloomberg. Prior to joining Stanford, she received her Bachelor's degree in Electrical Engineering and Computer Science at MIT and her PhD in Computer Science at UC Berkeley.
Autonomous Materials for Increased Agility and Endurance in Robots
Speaker: Prof. Dr. Robert Shepherd
Affiliation: Cornell University, USA
Date: November 07, 2025
Time & Location: 16:00 ; ETH HG F3
More info
Abstract
General purpose robots are an important technical challenge that would be enabling for agriculture, ocean health, space missions, disaster recovery, personal medical care, and many other uses. The proliferation of quadrotor and, now, quadrupedal robots are beginning to show the value of general utility, but still fail to achieve the mobility and manipulation dexterity, as well as efficiency and operational lifetime of animals. Animals have far better performance, but they are presently far too complicated to build synthetically. Nature balances the trade off between architectural complexity and energetic cost of construction better than we do. I will present context to complex material systems and autonomy, the concept of Autonomous Material Systems, as well as synthetic and living approaches towards this hierarchy and interconnectivity. The results indicate that distributing sensing, actuation, energy, and communication in robots has great advantages, but is likely insufficient without living materials. In this talk, I will focus on advances in the use of aqueous redox flow batteries as hydraulic fluids for high energy density underwater robots, arrays of powerful microscale combustion engines, and a bioelectronic interface with mycelia that can eventually be used to control these systems.
Bio
Robert Shepherd is the John F. Carr professor of engineering at Cornell University in the Sibley School of Mechanical & Aerospace Engineering. He received his B.S. (Material Science & Engineering), Ph.D. (Material Science & Engineering), and M.B.A. from the University of Illinois. At Cornell, he runs the Organic Robotics Lab (ORL: http://orl.mae.cornell.edu), that focuses on using methods of invention, including bioinspired design approaches, in combination with material science and mechanical design to improve machine function and autonomy. We rely on new and established synthetic approaches for soft material composites that create new design opportunities in the field of robotics. He is the recipient of an Air Force Office of Scientific Research Young Investigator Award, an Office of Naval Research Young Investigator Award, is a Senior Member of the National Academy of Inventors, and his lab's work has been featured in popular media outlets such as the BBC, Discovery Channel, and PBS's NOVA documentary series. He is an advisor to the American Bionics Project (americanbionics.org) which aims to make wheelchairs obsolete. He is also the co-founder of the Organic Robotics Corporation (DBA LLume; http://llume.io) which aims to digitally record the tactile interactions of humans and machines with their environment. He is also the co-founder of MAV-Unlimited, Inc. (https://mav-unlimited.com), which seeks to make mass customization of consumer goods possible via high throughput 3D printing.
Voluntary: Swiss Robotics Day in Lausanne
Speaker:
Affiliation:
Date: November 14, 2025
Time & Location: all day ; SwissTech Convention Center (Rte Louis Favre 2, 1024 Ecublens)
TBD
Speaker: Prof. Dr. Cosimo Della Santina
Affiliation: TU Delft, Netherlands
Date: November 28, 2025
Time & Location: 16:00 ; ETH HG G5
Unified Resilient Autonomy across Robot Configurations
Speaker: Prof. Dr. Kostas Alexis
Affiliation: Norwegian University of Science and Technology (NTNU), Norway
Date: December 05, 2025
Time & Location: 16:00 ; ETH HG G3
More info
Abstract
State-of-the-art autonomy methods remain fragmented with controllers, sensor fusion pipelines, and learning algorithms typically tailored to narrow robot morphologies and operating regimes. This specialization has historically been necessary to achieve operational results, but inevitably limits generalization and slows the pace of innovation. A common blueprint for autonomy is necessary. This talk outlines a vision toward Unified Resilient Autonomy that is applicable across diverse robot configurations, whether flying, aquatic, or ground systems. By pursuing a common autonomy architecture and leveraging the lessons learned from its broad evaluation in extreme conditions, we demonstrate resilient functionality that transfers across embodiments. The discussion will highlight both the underlying methods that enable this unification as well as concrete results from field testing in unconventional environments - such as subterranean settings, ship ballast tanks, and submarine bunkers.
Bio
Prof. Dr. Kostas Alexis is a Professor at the Department of Engineering Cybernetics at the Norwegian University of Science and Technology (NTNU), head of the Autonomous Robot Lab, and Director of the Norwegian Centre for Embodied AI. Together with his team, he conducts research on resilient robotic autonomy, exploring how autonomous systems can operate in high-risk, uncertain environments by presenting resourcefulness, robustness, and redundancy. Focusing on enhancing and safeguarding the autonomy capabilities of robotic systems, his research cuts across model-based optimization for control, sensor fusion, path planning, and learning algorithms for navigation. Prof. Alexis has served as Principal Investigator in major international grants both in Europe and the US, and was the PI and team lead of Team CERBERUS, winners of the DARPA Subterranean Challenge.
TBD
Speaker: Prof. Dr. Dino Accoto
Affiliation: KU Leuven, Belgium
Date: December 12, 2025
Time & Location: 16:00 ; ETH HG G5
Recent Past Talks
Multifunctional, Adaptive, and Sustainable Robotic Materials Towards General Utility
Prof. Dr. Hedan Bai (ETH Zurich, CH)
October 2025
Soft-Matter Engineering for Robotics and Wearables
Prof. Dr. Carmel Majidi (Carnegie Mellon University, USA)
October 2025
Molecular nanorobots: programming motility and cell interactions towards precision medicine
Prof. Dr. Tania Patino (Eindhoven University of Technology, Netherlands)
September 2025
Archives
- Spring 2025 - 14 talks on soft robotics, foundation models, and embodied AI
- More semesters coming soon…