Embedded Files
How can Computational and Numerical Modeling Enable the Next Generation of
Soft Robots?
Live Event: 20 October 2025 (1400-1800, GMT+8), Hangzhou International Expo Center, Hangzhou, China
Workshop Description
Soft robotics has been steadily advancing as a highly interdisciplinary field, often inspired by nature. As newer paradigms emerge on designing, actuating, and controlling soft bodies, what role can computational modeling of soft robots play? Can we create digital twins of our soft robot designs? Does computational modeling only assist as a validation tool, or can it even meaningfully inform design and fabrication choices? Can we move towards unified modeling frameworks while maintaining the focus on application and physics-specific computational techniques? What numerical models should one explore and adopt to better understand or improve the behavior of a soft robot? As the community seeks to enhance soft bodies' design, actuation, and control, computational modeling emerges as a critical tool. We believe that creating high-fidelity numerical models of soft robots can enhance the design-fabrication pipeline and provide new insights into the underlying physics of the robots' capabilities. Accurate computational models can provide a bridge to translate observed principles from biology to soft robots and from soft robots to biology. This workshop aims to facilitate an inter-disciplinary discussion between experts of different frameworks on how we can address the core challenges in soft robot modeling and comment on strategies to increase adoption and accessibility for such frameworks. The workshop also aims to help junior researchers starting out in the field to make better modeling choices by interacting with senior researchers with diverse experiences in robot modeling. By understanding the principles and rationale behind existing modeling approaches and the mathematics and physics behind those approaches, we seek to gain a comprehensive overview of current progress in computational modeling. This will help pinpoint which avenues to explore further to hopefully create digital twins of our soft robots, and also nurture a comprehensive discussion covering topics ranging from ground-up modeling to adoption of the right modeling techniques.
Workshop Description
Soft robotics has been steadily advancing as a highly interdisciplinary field, often inspired by nature. As newer paradigms emerge on designing, actuating, and controlling soft bodies, what role can computational modeling of soft robots play? Can we create digital twins of our soft robot designs? Does computational modeling only assist as a validation tool, or can it even meaningfully inform design and fabrication choices? Can we move towards unified modeling frameworks while maintaining the focus on application and physics-specific computational techniques? What numerical models should one explore and adopt to better understand or improve the behavior of a soft robot? As the community seeks to enhance soft bodies' design, actuation, and control, computational modeling emerges as a critical tool. We believe that creating high-fidelity numerical models of soft robots can enhance the design-fabrication pipeline and provide new insights into the underlying physics of the robots' capabilities. Accurate computational models can provide a bridge to translate observed principles from biology to soft robots and from soft robots to biology. This workshop aims to facilitate an inter-disciplinary discussion between experts of different frameworks on how we can address the core challenges in soft robot modeling and comment on strategies to increase adoption and accessibility for such frameworks. The workshop also aims to help junior researchers starting out in the field to make better modeling choices by interacting with senior researchers with diverse experiences in robot modeling. By understanding the principles and rationale behind existing modeling approaches and the mathematics and physics behind those approaches, we seek to gain a comprehensive overview of current progress in computational modeling. This will help pinpoint which avenues to explore further to hopefully create digital twins of our soft robots, and also nurture a comprehensive discussion covering topics ranging from ground-up modeling to adoption of the right modeling techniques.
Workshop Content and Objectives
Workshop Content and Objectives
This workshop aims to focus on several different aspects of computational modeling across various disciplines- differentiable simulations, biomechanical modeling, surgical robotics, computational fabrication, etc. We plan to bring together experts from across the domains of (soft) robot software and hardware modeling and computational graphics in order to create a multidisciplinary discussion on different modeling techniques, which we believe will help the audience understand where the field is at, and how we can use modeling as a tool to advance soft robot design and research. A large variety of modeling methods are utilized for different applications, and for researchers just starting their journey in the field, it can be difficult to make a decision on which modeling framework may best suit their needs, with an invariably steep learning curve associated with most frameworks. Through this workshop's elements, which include expert talks as well as student presentations and demos, we aim to brainstorm a roadmap as well as provide relevant information needed to pick up and implement modeling approaches pertinent to different research problems in soft robotics. Moreover, since biological inspiration plays such a central role in soft robot design and control, how can we make these modeling tools more faithful regarding their replication of nature? This workshop aims to provide an interactive platform where modeling experts share their insights on the techniques they developed, interspersed with discussions on the perceived advantages and points of improvement in terms of optimization, accuracy, speed, etc., for the different techniques. We aim to incorporate perspectives of researchers across the spectrum of career stages to increase insights on which issues need to be addressed to advance soft robot modeling. Some key questions we aim to answer are:
1) What are the core challenges of computational and numerical modeling in soft robotics?
2) How do we create accurate digital twins of soft robots?
3) What motivated the design choices behind the existing frameworks?
4) How do we bridge the differences between different methods and approaches, and do we need to do that?
5) Can there be a 'perfect' simulator?
and others..
Relevant research topics include: biological modeling, digital twins, numerical techniques (finite-element, Cosserat theory), sim2real gap, differentiable simulations, design optimization, computational fabrication and graphics, and many more.
We aim to achieve several outcomes from this workshop, including but not limited to:
This workshop aims to focus on several different aspects of computational modeling across various disciplines- differentiable simulations, biomechanical modeling, surgical robotics, computational fabrication, etc. We plan to bring together experts from across the domains of (soft) robot software and hardware modeling and computational graphics in order to create a multidisciplinary discussion on different modeling techniques, which we believe will help the audience understand where the field is at, and how we can use modeling as a tool to advance soft robot design and research. A large variety of modeling methods are utilized for different applications, and for researchers just starting their journey in the field, it can be difficult to make a decision on which modeling framework may best suit their needs, with an invariably steep learning curve associated with most frameworks. Through this workshop's elements, which include expert talks as well as student presentations and demos, we aim to brainstorm a roadmap as well as provide relevant information needed to pick up and implement modeling approaches pertinent to different research problems in soft robotics. Moreover, since biological inspiration plays such a central role in soft robot design and control, how can we make these modeling tools more faithful regarding their replication of nature? This workshop aims to provide an interactive platform where modeling experts share their insights on the techniques they developed, interspersed with discussions on the perceived advantages and points of improvement in terms of optimization, accuracy, speed, etc., for the different techniques. We aim to incorporate perspectives of researchers across the spectrum of career stages to increase insights on which issues need to be addressed to advance soft robot modeling. Some key questions we aim to answer are:
1) What are the core challenges of computational and numerical modeling in soft robotics?
2) How do we create accurate digital twins of soft robots?
3) What motivated the design choices behind the existing frameworks?
4) How do we bridge the differences between different methods and approaches, and do we need to do that?
5) Can there be a 'perfect' simulator?
and others..
Relevant research topics include: biological modeling, digital twins, numerical techniques (finite-element, Cosserat theory), sim2real gap, differentiable simulations, design optimization, computational fabrication and graphics, and many more.
We aim to achieve several outcomes from this workshop, including but not limited to:
Help attendees leave with a better understanding of different computational modeling techniques relevant to soft robotics, equipped with the necessary knowledge to select tools that can best help their research
By giving students a chance to present software demos, we foresee stimulating discussion on a wide variety of topics in soft robot modeling and giving junior researchers ample time to interact with fellow junior researchers as well as get quality feedback from senior researchers to fuel the innovation of novel ideas in modeling
Expand the soft robotics modeling community by introducing new voices and expertise, while promoting diversity in geography, gender, and research backgrounds
We invite students and early career researchers working on modeling for soft robotics to submit a 2-page extended abstract, with a minimum of 2 figures and optionally a demo of their software/simulations. Accepted submissions will get to present a poster, and selected demos will give spotlight talks during the workshop. The demo can be a screen-recording walk-through of a small simulation of soft robotics research (example: FEM modeling of a soft robot or a shape-optimization algorithm), along with a link to a GitHub repo if you are writing/utilizing custom code. Please check out the website page on the call for contributions to find out more.
Important Dates
Workshop Date: 20 October 2025
Abstract Submission Deadline: 15 August 2025
Abstract and Demo Acceptance Notification: 21 August 2025
Point of Contact for Any Queries: adamya@u.nus.edu
Topics of Interest for submissions (not exhaustive, please email us if you are unsure):
1) Simulations and Animations for Soft Robots
2) Trajectory Optimization
3) Computational Fabrication
4) Differentiable Simulations
5) Computational Design
6) Biological Modeling
7) Numerical Methods (FEM, Cosserat etc.)
8) Learning-based and Model Predictive Control
9) Kinematics and Dynamics Modeling
10) Musculoskeletal and Biomechanical Modeling
Confirmed Speakers
Confirmed Speakers
Max Planck Institute for Intelligent Systems
(Musculoskeletal Modeling)
University of Alberta
(Surgical Robotics)
TU Delft
(Control and Kinematics Modeling)
Diversity, Equity and Inclusion Statement
As the organizers of this workshop, we are strongly committed to promoting diversity at every level and through every aspect of the workshop. We value and prioritize representation from historically underrepresented groups. Our speaker lineup includes a diverse group of experts from various disciplines, career stages, and institutions worldwide, with a strong focus on gender balance. We actively seek to engage participants from different regions and backgrounds, fostering an inclusive platform for interdisciplinary discussions that advance our community. We are also actively collaborating with organizations linked to underrepresented groups to solicit broader participation.
Our workshop at IROS 2025 will adhere to the RAS guidelines for workshops and tutorials, as well as the IEEE RAS Code of Conduct, ensuring a respectful and inclusive environment for all participants.
Organizing Team
ETH Zürich/Max Planck Institute for Intelligent Systems
ETH Zürich
ETH Zürich
National University of Singapore
National University of Singapore
National University of Singapore
ETH Zürich
Funding Acknowledgements
Singapore Ministry Of Education
A*STAR
Italian Ministry of Foreign Affairs and International Cooperation
Swiss National Science Foundation
ETH AI Center
Max Planck-ETH Center for Learning Systems
Sponsors
Sponsors
TBD
Page updated
Google Sites
Report abuse