An increasing number of robots are made of soft materials. Because they are inherently safe to interact with, soft robots have been widely used in human-robot interactions. Moreover, due to their flexibility, they adapt to their environment, requiring less precision in certain tasks like grasp planning.

Nonetheless, soft robotic state estimation has its unique challenges. In general, soft robots are underactuated and have infinitely many degrees of freedom. They undergo large deformations and hence display a highly nonlinear hyperelastic behaviour. Furthermore, because many applications involve contact, we require state estimation algorithms to scale well and perform robustly in contact-rich interactions with humans or environments. To balance precision and timing requirements, novel proprioceptive sensing and Modeling are needed to estimate a soft robot’s state efficiently and effectively. Real-time state estimation also paves the way towards closed-loop control of soft robots.

This workshop aims to exchange new knowledge, discover inspiring trends, and advance integrated software and hardware techniques for soft robot sensing and the estimation of their internal states, such as geometric shape and properties, strain- and stress-related quantities in quasi-static settings, and velocities and accelerations in dynamic settings. Our goal is to connect communities and stimulate collaborative research.

• Edward H. Adelson, MIT, (Tactile Sensing, Learning-based Sensing)
• Robert Shepherd, Cornell University (Soft Robotics and Soft Sensors)
• Oliver Brock, Technische Universität Berlin (Biology-Inspired & Learning-based Robotics)
• Daniele Panozzo, NYU (Graphics & Simulation for Soft Robotics)
• Rebecca Kramer-Bottiglio, Yale University (Soft Robotics, Multifunctional Materials)
• Miguel Otaduy, Universidad Rey Juan Carlos (Soft Body Simulation, Contact Modeling)
• Mike Tolley, University of California, San Diego (Soft Robotics and Self-Assembly)

We invite researchers working on related topics to submit abstracts or extended abstracts (no longer than 4 pages in IROS paper format, including references) that can contribute to this workshop.

Desired Works could:

• identify novel state estimation theories and algorithms for robots with parts made of soft materials,
• discuss state estimation in the context of applications in soft robotics, human-robot interaction, or wearable devices,
• demonstrate real-time state estimation with limited precision for closed-loop or accurate state estimation for open-loop control,
• describe novel integrated hard- and software prototypes of proprioceptive soft robots,
• review and benchmark various methods proposed by different communities (e.g., robotics, HCI, HRI) with the ultimate goal to enhance the mutual understanding of challenges and opportunities related to this workshop.

TBD

The organizers are currently guest editing a special issue on the same research topic (https://www.frontiersin.org/research-topics/19049/soft-robot-state-estimation). The potential authors would also be invited to participate in this workshop. Authors of the selected abstracts would be invited to submit the full version of their works to this special issue.

• Soft body state Modeling and estimation
• Soft sensor design
• Shape and tactile sensing
• Machine learning for soft robotics
• Soft robot simulation
• High-resolution soft body proprioception
• Contact force/torque estimation
• Joint proprioception and tactile sensing
• Real-time state estimation for soft bodies
• New mechanisms or principles for soft body state estimation
• Simulation and differentiable simulation for soft body state estimation
• Learning-based tactile sensing
• Learning-based proprioception
• Self-calibration for soft robot state estimation
• Soft robot co-design for state estimation and closed-loop control
• Perspectives on challenges and open questions
• Benchmarking experiments for comparing sensing techniques and state estimation algorithms

TBD

• Chen Feng, NYU, cfeng at nyu dot edu
• Wenzhen Yuan, CMU, wenzheny at andrew dot cmu dot edu
• Huichan Zhao, Tsinghua University, zhaohuichan at mail dot tsinghua dot edu dot cn
• Moritz Bächer, Disney Research, moritz.baecher at disney dot com