Category Archives: Projects

Pages of the research projects

Medical Application of Origami and Soft Robotic Systems

Our lab’s research in origami‐inspired robotics extends into the medical arena through collaborations with the University of Pennsylvania Hospital and the Children’s Hospital of Philadelphia, including partnerships with the Departments of Cardiology and Plastic Surgery. We also work closely with faculty such as Prof. Jordan Raney (MEAM) and Prof. Flavia Vitale (Penn’s Center for Neuroengineering & Therapeutics). These efforts focus on reconfigurable implants, artificial muscles, and soft actuators for medical devices.

A central goal is to develop origami-inspired soft actuators that function as artificial muscles, leveraging advanced fabrication techniques to enable compact, flexible, and highly robust motion. By incorporating principles such as multistability and bistability from origami, we design actuator systems that can be easily reconfigured yet remain strong enough to perform clinically relevant tasks. Our lab also leads work on mechanical characterization of origami-inspired tubular structures for use as reconfigurable implants, aiming to reduce surgical invasiveness by creating implantable devices (e.g., heart or bile duct stents) that can be adjusted noninvasively.

In parallel, our lab is developing algorithms and interactive design tools for kinematic mechanisms, with the goal of enabling the rapid and affordable creation of customized orthotics and prosthetics. These computational tools are intended to streamline the design process for patient-specific assistive devices, broadening access to personalized care.

MORF in Medical Applications
Our MORF (Magnetic Origami Reprogramming and Folding) System—initially developed for general reconfigurable devices—has proven especially promising for medical stents. Recently, in Penn’s Y-Prize competition, the 2025 winners “Stentix” proposed a magnetically reconfigurable biliary stent based on MORF. By using magnetic forces, this stent can be adjusted in position and diameter from outside the body, helping maintain bile flow without repeat endoscopies.

Related Publications

  1. C. Kim, L. Yang, A. Anbuchelvan, R. Garg, N. Milbar, F. Vitale, and C. Sung, “Origami-Inspired Bistable Gripper with Self-Sensing Capabilities,” 2024 IEEE-RAS 7th International Conference on Soft Robotics (RoboSoft), San Diego, CA, USA, 2024 (Accepted)
  2. B. Leung, G. Unger, S. Escorza, J. Chen, M. Fogel, and C. Sung, “Mechanical Characterization of an Origami-Inspired Multistable Tube for Reconfigurable Implants,” Poster at the Biomedical Engineering Society’s (BMES) Annual Meeting, 2023
Mechanical Characterization of an Origami-Inspired Multistable Tube for Reconfigurable Implants,

Leung, Brianna; Unger, Gabriel; Escorza, Saúl; Chen, Jonathan; Fogel, Mark; Sung, Cynthia

Mechanical Characterization of an Origami-Inspired Multistable Tube for Reconfigurable Implants, (Conference)

Biomedical Engineering Society’s (BMES) Annual Meeting, 2024.

(BibTeX)

Origami-Inspired Bistable Gripper with Self-Sensing Capabilities

Kim, Christopher; Yang, Lele; Anbuchelvan, Ashwath; Garg, Raghav; Milbar, Niv; Vitale, Flavia; Sung, Cynthia

Origami-Inspired Bistable Gripper with Self-Sensing Capabilities (Conference)

IEEE-RAS International Conference on Soft Robotics (Robosoft), 2024.

(Abstract | BibTeX | Links: )

Current Personnel

  • Gabriel Unger (MEAM PhD)
  • Kayleen Smith (MEAM Undergrad)
  • Kylie Autullo (MEAM Undergrad)

Past Personnel

  • Christopher Kim (MEAM PhD)
  • Brianna Leung (BE Undergrad)
  • Harita Trivedi (BE Undergrad)
  • Alec Lanter (MEAM Master's)
  • Louis Beardell (University of Michigan, BE Visitor)
  • Serena Carson (ROBO Master's)

Acknowledgments

This work was supported in part by the Johnson & Johnson WiSTEM2D program, by the Children’s Hospital of Philadelphia (CHOP) Cardiac Innovation program, by the Penn Health-Tech program, and by the Penn Center for Undergraduate Research and Fellowships. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of funding source.

SALP: Salp-inspired Approach to Low-Energy Propulsion

Jet propulsion is a locomotion mode commonly found in biological swimmers, including cephalopods and tunicates such as squids, cuttlefish, and salps. We are developing a soft salp-inspired robotic system to study mechanisms that produce greater locomotion agility and energetic efficiency.

Salps are barrel-shaped marine invertebrates that swim via jet propulsion. They move forward by rapidly changing the volumes of their body cavity, drawing water into their muscular mantle cavity through the front aperture, and then expelling it under high pressure through the rear funnel. We aim to leverage the unique biomechanics of salps to inform the development of energy-efficient, maneuverable underwater robots capable of environmental sensing in complex marine environments.

Salps can swim either as solitary jet-propelled individuals or while physically connected in a multi-jet colony, commonly known as a “salp chain”. Inspired by salps, we develop the SALP (Salp-inspired Approach to Low-energy Propulsion) robot, a soft underwater robot that swims via jet propulsion similarly to a biological salp.

Version 1: Origami Swimmer

Traditional soft robots use flexible materials for shape change but require complex fabrication. To simplify this, we used an origami-inspired design, which allows the robot to fold from flat sheets into 3D shapes, making it easier to store, transport, and assemble in just a few hours. The robot mimics squid locomotion using the origami magic ball pattern, which transforms between an ellipsoid and a sphere. This enables it to expand and contract, pulling in and expelling water to create a jet for propulsion. A tendon mechanism in its spine controls its length. Ongoing effects focus on evaluating different configurations’ effect (e.g. adding a front nozzle) on the performance of the robot and understanding the underlying dynamics of it.

Mechanism demo

Robot swimming

Version 2: Multi-Robot SALP Robot Platform for Long-Term Distributed Sensing

We have optimized the jetting robot design to enable higher thrust and lower drag. The SALP robot is now modular in the sense that the robots can be manually attached to each other in different physical arrangements to study the effect of multi-robot interactions on locomotion performance. Physically connected SALP chains coordinate their jets to achieve various propulsion modes. We are interested in investigating how physical arrangement and jet coordination between two robots affect the swimming performance and energy efficiency of a two-SALP robotic system. We aim to gain insights into the potential hydrodynamic benefits of multi-jet propulsion by exploring how different coordination strategies influence the surrounding flow environment.

Leveraging Fluid-Structure Interactions for Efficient Control in Geophysical Flows

Micro-vehicles are cost-effective platforms for robotics and automation, excelling in maneuverability and adaptability in diverse environments. However, their lightweight and limited computational capacity pose control challenges. Using our underwater platform, we aim to understand fluid-structure interactions to enhance design and control, resulting in more efficient micro-vehicles with extended lifespans. This effort is a collaborative project focusing on fluid dynamics, control theory, and reconfiguration planning. The project aims to leverage environmental forces for power efficiency, investigating morphological adaptations and passive transport properties. It seeks to synthesize motion control strategies considering inertial effects and fluid-structure interactions while exploring efficiency trade-offs. Ultimately, it aims to enhance micro-autonomous vehicles’ capabilities for long-term operations and future large-scale deploy.

Related Publications

Effect of Jet Coordination on Underwater Propulsion with the Multi-Robot SALP System

Yang, Zhiyuan; Zhang, Yipeng; Herbert, Matthew; Hsieh, M. Ani; Sung, Cynthia

Effect of Jet Coordination on Underwater Propulsion with the Multi-Robot SALP System (Conference)

8th IEEE-RAS International Conference on Soft Robotics (RoboSoft 2025), Forthcoming.

(Abstract | BibTeX | Links: )

Drag coefficient characterization of the origami magic ball

Chen, Guanyu; Chen, Dongsheng; Weakly, Jessica; Sung, Cynthia

Drag coefficient characterization of the origami magic ball (Proceedings Article)

In: ASME International Design Engineering Technical Conferences and Computers and Information in Engineering Conference (IDETC/CIE), pp. DETC2023-117182, 2023.

(Abstract | BibTeX | Links: )

Origami-inspired robot that swims via jet propulsion

Yang, Zhiyuan; Chen, Dongsheng; Levine, David J.; Sung, Cynthia

Origami-inspired robot that swims via jet propulsion (Journal Article)

In: IEEE Robotics and Automation Letters, vol. 6, no. 4, pp. 7145-7152, 2021.

(Abstract | BibTeX | Links: )

Current Personnel

  • Dongsheng Chen (MEAM PhD)
  • Zhiyuan (Annie) Yang (MEAM PhD)
  • Ryan Stanford (MEAM Undergrad)
  • Benedict Onyekwe (ROBO Master's)
  • Jingshuo Li (MEAM Master's)
  • Neel Mulay (MEAM Master's)

Past Personnel

  • Adithya Selvakumar (ESE Undergrad)
  • Guanyu Chen (MEAM Master's)
  • Yipeng Zhang (MEAM Master's)
  • Yunyi Chu (MEAM Master's)
  • Zhiyuan (Annie) Yang (MEAM, MCIT Master's)

Acknowledgments

The project “Leveraging Fluid-Structure Interactions for Efficient Control in Geophysical Flows” is in collaboration with Ani Hsieh’s lab from University of Pennsylvania, Eric Forgoston’s lab from Montclair State University, and Philip Yecko’s lab from The Cooper Union.

These projects have been supported by the National Science Foundation (NSF) Grant No. 2121887 and the Office of Naval Research (ONR) award #N00014-23-1-2068. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of funding source.

Overview diagram for Kinegami system

Kinegami: Computational Design of Kinematic Mechanisms

This project aims to develop computational pipelines for users to quickly and cheaply design and construct mechanisms from kinematic specifications. Arms, legs, and fingers of animals and robots are all examples of “kinematic chains” – mechanisms with sequences of joints connected by effectively rigid links. We create end-to-end design algorithms and interactive editing software for kinematic “skeletons” that can be fabricated as origami or 3D printed structures. This is part of a larger effort within the lab to provide tools for rapid prototyping and fabrication of custom robots and mechanisms.

Overview diagram for Kinegami system

Compositional Design of Tubular Structures

We construct kinematic chains and trees as tubular structures designed as compositions of rotational and translational modules. The methods are built upon a library of parameterized designs for revolute (rotating) joints, prismatic (sliding) joints, and rigid links. We have designed a library of modules constructed from origami for kinematic chains. Currently, we are working on 3D printed module designs, including branching modules to extend our work to kinematic trees.

Algorithms

Our algorithms automatically design kinematic chains and trees with given degrees of freedom. Given a sequence of axes of motion (lines in 3D space along which a revolute joint rotates or a prismatic joint translates), our algorithms calculate a position and orientation along each axis such that joints can be sequentially connected by tubular links. The core idea is to convert the design problem into a planning problem for module centerline paths. Since a tube cannot bend more sharply than its own radius, the paths have a minimum turning radius, making this a Dubins planning problem. The algorithms space joints far enough apart and orient them appropriately such that collision-free Dubins paths exist connecting them.

Human-in-the-Loop Design Tools

We are creating fully interactive design software to enable humans to create kinematic chains and trees with assistance from our algorithms, requiring no coding or engineering expertise. We currently have a python repository for creating and editing tubular kinematic chains, visualizing how they can move, and exporting origami crease patterns to construct them: see our github repo for more details.

Dynamical Robots

(In collaboration with Kod*lab)

Our library of tubular origami module patterns enables rapid, cheap, semi-automated prototyping of dynamical robots with high power density. To demonstrate this, we are building the Dynamic Origami Quadruped (DOQ), an untethered mesoscale robot capable of walking, bounding, and pronking gaits. The light weight of the origami tubes enables the robot mass to be about 50% actuators.

Resources

Python code for creating and editing tubular origami kinematic chains (from our 8OSME paper, 2024): https://github.com/SungRoboticsGroup/KinegamiPython

MATLAB code for creating tubular origami kinematic chains (from our 2023 T-RO paper): https://github.com/SungRoboticsGroup/Kinegami

Instructional example videos for folding tubular origami modules. These examples have 4 sides and are made from .005″ thick PET plastic film. The crease lines are laser etched at 25 PPI, and mountain-valley coloring is hand-drawn in pen.

Related Publications

Algorithmic Design of Kinematic Trees Based on CSC Dubins Planning for Link Shapes

Feshbach, Daniel; Chen, Wei-Hsi; Xu, Ling; Schaumburg, Emil; Huang, Isabella; Sung, Cynthia

Algorithmic Design of Kinematic Trees Based on CSC Dubins Planning for Link Shapes (Conference)

Workshop on the Algorithmic Foundations of Robotics (WAFR), 2024.

(Abstract | BibTeX | Links: )

Reparametrization of 3D CSC Dubins' Paths Enabling 2D Search

Xu, Ling; Baryshnikov, Yuliy; Sung, Cynthia

Reparametrization of 3D CSC Dubins' Paths Enabling 2D Search (Conference)

Workshop on the Algorithmic Foundations of Robotics (WAFR), 2024.

(Abstract | BibTeX)

Kinegami: Open-source Software for Creating Kinematic Chains from Tubular Origami

Feshbach, Daniel; Chen, Wei-Hsi; Koditschek, Daniel E.; Sung, Cynthia

Kinegami: Open-source Software for Creating Kinematic Chains from Tubular Origami (Conference)

8th International Meeting on Origami in Science, Mathematics, and Education (8OSME), 2024.

(Abstract | BibTeX | Links: )

Robogami Reveals the Utility of Slot-Hopper for Co-Design of DOQ’s Body and Behavior

Chen, Wei-Hsi; Caporale, J. Diego; Koditschek, Daniel E.; Sung, Cynthia

Robogami Reveals the Utility of Slot-Hopper for Co-Design of DOQ’s Body and Behavior (Workshop)

ICRA 2024 Workshop on Co-design in Robotics: Theory, Practice, and Challenges, 2024.

(BibTeX | Links: )

Bio-inspired quadrupedal robot with passive paws through algorithmic origami design

Chen, Wei-Hsi; Qi, Xueyang; Feshbach, Daniel; Wang, Stanley J.; Kuang, Duyi; Full, Robert; Koditschek, Daniel; Sung, Cynthia

Bio-inspired quadrupedal robot with passive paws through algorithmic origami design (Workshop)

7th IEEE-RAS International Conference on Soft Robotics (RoboSoft) Workshop: Soft Robotics Inspired Biology, 2024.

(BibTeX | Links: )

DOQ: A Dynamic Origami Quadrupedal Robot

Chen, Wei-Hsi; Rozen-Levy, Shane; Addison, Griffin; Peach, Lucien; Koditschek, Daniel E.; Sung, Cynthia R.

DOQ: A Dynamic Origami Quadrupedal Robot (Workshop)

ICRA Workshop on Origami-based Structures for Designing Soft Robots with New Capabilities, 2023.

(BibTeX)

Kinegami: Algorithmic Design of Compliant Kinematic Chains From Tubular Origami

Chen, Wei-Hsi; Yang, Woohyeok; Peach, Lucien; Koditschek, Daniel E.; Sung, Cynthia R.

Kinegami: Algorithmic Design of Compliant Kinematic Chains From Tubular Origami (Journal Article)

In: IEEE Transactions on Robotics, vol. 39, iss. 2, pp. 1260-1280, 2023, (Honorable mention for 2023 IEEE Transactions on Robotics King-Sun Fu Memorial Best Paper Award).

(Abstract | BibTeX | Links: )

Current Personnel

  • Wei-Hsi Chen (ESE Postdoc)
  • Daniel Feshbach (CIS PhD)
  • Andy Wang (CIS Undergrad)
  • Daniel Lin (ESE Undergrad)
  • Emil Schaumburg (CMPE Undergrad)
  • Jeffery Oduman (CIS Undergrad)
  • Raymond Feng (DMD Undergrad)
  • Zachary Leong (DMD Undergrad)
  • Samhitha Vedire (ROBO Master's)

Past Personnel

  • Isabella Huang (ESE Undergrad)
  • Ling Xu (MEAM Undergrad)
  • Alex Chi (Germantown Friends School Visitor)
  • Eric Huang (Peddie School Visitor)
  • Lilianna Sand (Central High school Visitor)

Acknowledgments

This project has been supported by the National Science Foundation (NSF) under grants 2322898 and 1845339, and by the Army Research Office (ARO) under the SLICE Multidisciplinary University Research Initiatives Program grant W911NF1810327. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of funding source.

Self-Sensing Actuators

Overview

This project thrust introduces self-sensing actuators that merge actuation and sensing into a single component, enabling reductions in the size and weight of soft robots. Three self-sensing actuators—the I-cord knitted SMA (ICKS) actuator, the custom PET solenoid actuator from the Facial Reanimation and Bistable Gripper project, and a miniaturized tunable stiffness solenoid actuator—will showcase the self-sensing actuation capability.

Videos

I-cord Knitted SMA (ICKS) Actuator
Origami-Inspired Bistable Gripper with the Custom PET Solenoid Actuator

Publications

Origami-Inspired Bistable Gripper with Self-Sensing Capabilities

Kim, Christopher; Yang, Lele; Anbuchelvan, Ashwath; Garg, Raghav; Milbar, Niv; Vitale, Flavia; Sung, Cynthia

Origami-Inspired Bistable Gripper with Self-Sensing Capabilities (Conference)

IEEE-RAS International Conference on Soft Robotics (Robosoft), 2024.

(Abstract | BibTeX | Links: )

Fabrication and characterization of I-cord knitted SMA actuators

Kim, Christopher; Chien, Athena; Tippur, Megha; Sung, Cynthia

Fabrication and characterization of I-cord knitted SMA actuators (Conference)

IEEE International Conference on Soft Robotics (RoboSoft), 2021.

(Abstract | BibTeX | Links: )

People

  • Christopher Kim
  • Kylie Autullo
  • Lele Yang
  • Brianna Leung
  • Athena Chien
  • Megha Tippur

Acknowledgement

Support for this project has been provided in part by the National Science Foundation (NSF) grant #EEC-1659190, by the Johnson \& Johnson WiSTEM2D program, by the University of Pennsylvania through the Center for Precision Engineering for Health, and by the Edwin and Fannie Gray Hall Center for Human Appearance Research and Education Fund. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and and do not necessarily reflect the views of funding source.