Category Archives: Projects

Pages of the research projects

Artistic Non-Inertial Tracer (ANT): an Educational Kit for a 3-Link Origami Slithering Robot

Daniel Feshbach, Alex Chi, Eric Huang, Diedra Krieger, Cynthia Sung

The abundance of connections between art and engineering are opportunities to engage artistically inclined students who may not think of themselves as interested in robotics, and to inspire engineering-inclined students to express themselves artistically. This work presents a tool kit and lesson plan for a hands-on introductory robotics activity centering how art and engineering influence each other. The Artistic Non-Inertial Tracer (ANT) is a three-link robot made of origami, which contacts the ground through markers that trace patterns as it slithers around. The design is capable of forward, turning, backward, and diagonal motion. Using origami for the body highlights the influence of art on engineering, and the gait traces are a visually interesting output of the engineering. The kit uses inexpensive components totaling about $26/robot USD (servos, origami, and an Arduino Nano) common in hobbyist communities so that learning how to use these components can empower further exploration of actuated art. Our associated lesson plan engages participants in the assembly and control of ANT and contextualizes it within a broader overview of robot system components and the use of origami in engineering. We successfully ran the lesson as a 3-hour outreach workshop at a local arts organization.

Artistic Non-Inertial Tracer (ANT): an Educational Kit for a 3-Link Origami Slithering Robot

Feshbach, Daniel; Chi, Alex; Huang, Eric; Krieger, Diedra; Sung, Cynthia

Artistic Non-Inertial Tracer (ANT): an Educational Kit for a 3-Link Origami Slithering Robot (Conference)

2025 ASEE Annual Conference and Exposition, American Society for Engineering Education Forthcoming.

(Abstract | BibTeX | Links: )

Fabrication and Code Files

Instructional Materials

Gait Videos

Gaits are written as tuples (amplitude, joint 1 center angle, joint 2 center angle, cycle duration, number of cycles) with angles in degrees and cycle duration (period) in seconds.

TRUSSES: Temporarily, Robots Unite to Surmount Sandy Entrapments, then Separate

This project is a collaboration with research groups at Penn (Mark Yim, Daniel Koditschek, Douglas Jerolmack) and at USC (Feifei Qian). This aim for this project is to develop methods for teams of robots to jointly overcome environmental hazards on the Moon by attaching to each other to form larger and more stable, maneuverable structures. The robots will use their interactions with the ground to form a map of safe and risky terrain, attach to each other as support when the ground traversal risk is high, move in a coordinated fashion once joined, and, once the maneuver has been successfully completed, separate to continue their original individual missions.

Objectives

The TRUSSES project will develop algorithms that provide robots with two main capabilities: 1) estimate robot-to-regolith interactions in order to plan safe maneuvers, and 2) plan truss formations and coordinated motions for robots to push and pull each other to safe locations. The system will be evaluated to verify risk estimation, risk avoidance, risk mitigation, and recovery from failure.

Current Personnel

  • Rachel Holladay (MEAM Postdoc)
  • Shivangi Misra (ESE PhD)
  • William Hoganson (ROBO Staff)
  • Eric Wang (CIS Undergrad)
  • Lori Brown (CIS Undergrad)
  • Neha Peddinti (CIS Undergrad)
  • Wilson Hu (CIS Undergrad)
  • Benjamin Aziel (MEAM Master's)
  • Paul Young (MEAM Master's)

Acknowledgments

The work is supported by NASA’s Lunar Surface Technology Research (LuSTR) 2023 program. 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 this organization.

Tunable Stiffness in Soft Robots

The emergence of smart actuators with tunable material properties has empowered roboticists to create systems capable of transitioning between rigid and soft operational modes. This adaptability allows robots to operate with high precision and enhanced payload capacity in their rigid state, while offering greater safety and flexibility when soft especially in the presence of unpredictable disturbances. We design these adaptive robotic systems and develop corresponding control algorithms to showcase the versatility and potential of this new class of reconfigurable robots.

Fig 1: Collage of projects in the research thrust: Tunable Stiffness in Soft Robots

Coiled Spring Actuator

We have developed several novel tunable-stiffness actuators as part of this research thrust, one of which is the coiled-spring actuator. It is inspired by the mechanics of nested elastic rings, wherein the effective bulk stiffness can be modulated by varying the number of elastic coiled layers. This mechanism enables near-linear stiffness tuning, achieved through electro-mechanical control. The system allows for precise, programmable stiffness adjustments, and our prototype has demonstrated a tunability range of up to 20-fold. We have also constructed a non-dimensional mechanics model for the coiled spring actuator which extends to all such mechanisms of different dimensions and materials. The local stiffness changes from these actuators induce corresponding deformations in a compliant, segmented manipulator constructed as a tower of multiple such modules.

Pneumatic Dual-Bellows Actuator

Another novel tunable stiffness actuator developed in our lab is able to achieve a stiffness gain of  1.43 times (1332 N/m to 1913 N/m) without needing an external air source or valve. The design consists of an air chamber bellows and spring bellows connected to each other in an air-tight manner. Stiffness modulation in the spring bellows is achieved by altering the volume of the air chamber bellows. Due to large achievable stiffnesses, this actuator is suitable for integration in soft robots that are needed to demonstrate dynamic and adaptable behavior.

Hierarchical Algorithms to Optimize Soft Manipulator Mechanics

Stiffness control algorithms are needed for soft manipulators to be able to take effective advantage of embedded novel stiffness actuators. To address this, we have developed a hierarchical policy for stiffness control for a class of soft segmented manipulators. The stiffness changes induce desired deformations in each segment, thereby influencing the manipulator’s end-effector position. The algorithm can be run as an online controller to influence the manipulator’s stable states or offline as a design algorithm to optimize stiffness distributions.

Coupled Learning in Elastic Networks

We are also collaborating with Lauren Altman and Doug Durian in the Physics Department, School of Arts and Sciences, University of Pennsylvania to build an elastic network for coupled learning. This technique tunes the properties of individual elastic elements in the network to achieve specific outcomes, like applying the right force or strain to an output edge. These mechanical networks could be scaled and automated to tackle more complex tasks, opening the door to a new kind of smart metamaterial.

Related Publications

Online Optimization of Soft Manipulator Mechanics via Hierarchical Control

Misra, Shivangi; Sung, Cynthia

Online Optimization of Soft Manipulator Mechanics via Hierarchical Control (Conference)

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

(Abstract | BibTeX | Links: )

Design and Characterization of a Pneumatic Tunable-Stiffness Bellows Actuator

Chen, Rongqian; Kwon, Jun; Chen, Wei-Hsi; Sung, Cynthia

Design and Characterization of a Pneumatic Tunable-Stiffness Bellows Actuator (Conference)

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

(Abstract | BibTeX | Links: )

Design and Control of a Tunable-Stiffness Coiled-Spring Actuator

Misra, Shivangi; Mitchell, Mason; Chen, Rongqian; Sung, Cynthia

Design and Control of a Tunable-Stiffness Coiled-Spring Actuator (Conference)

IEEE International Conference on Robotics and Automation (ICRA), 2023.

(Abstract | BibTeX | Links: )

Forward kinematics and control of a segmented tunable-stiffness 3-D continuum manipulator

Misra, Shivangi; Sung, Cynthia

Forward kinematics and control of a segmented tunable-stiffness 3-D continuum manipulator (Conference)

IEEE International Conference on Robotics and Automation (ICRA), 2022.

(Abstract | BibTeX | Links: )

Current Personnel

  • Wei-Hsi Chen (ESE Postdoc)
  • Shivangi Misra (ESE PhD)
  • Allen Zhou (MEAM Master's)
  • Ruizhe Wang (ESE Master's)

Acknowledgments

Support for these projects has been provided by NSF Grant No. 1845339, the Johnson & Johnson WiSTEM2D Scholars Program, and the Army Research Office (ARO) under MURI award #W911NF1810327. Any opinions, findings, conclusions, or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of these organizations.

Template-Generated Robots

Overview

Designing robots requires a careful balance between their physical structure and behavior, as modifications in one area typically necessitate changes in the other. To enhance this design process, we introduce a co-design strategy that integrates modularization and abstraction, specifically focusing on the Dynamic Origami Quadruped (DOQ). This method aims to simplify design decisions by minimizing the interactions between components, enabling the creation of versatile robots while reducing the complexity of choices required from designers.

Our approach is grounded in two key concepts. The first involves utilizing dynamical systems templates to abstract essential locomotion dynamics, providing vital constraints that inform kinematics and actuation. The second concept is the Robogami (Kinegami) prototyping technique, which translates high-level specifications into feasible fabrication plans. Together, these frameworks streamline the design process, allowing designers to concentrate on overarching goals while automatically generating the specifics needed for successful prototypes. This work lays the groundwork for future innovations in robotics design.

Related Publications

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)

Current Personnel

  • Wei-Hsi Chen (ESE Postdoc)
  • Dong Wook Kim (ROBO Staff)
  • Henry Westfall (MEAM Undergrad)
  • Lindsay Fabricant (MEAM, Wharton Undergrad)
  • Sophie Abramovitz (MEAM Undergrad)
  • Vanessa Gong (MEAM undergrad)
  • Dhyey Shah (ROBO Master's)
  • Solomon Gonzalez (MEAM Master's)

Acknowledgments

This project has been supported by the National Science Foundation under grants 2322898 and 1845339, by the Army Research Office under Grant W911NF2410090 and 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 the National Science Foundation or the Army Research Office.

Programmable Matter

This work explores innovative ways to enhance the control, reliability, and fabrication of programmable matter systems—systems with the ability to change their shape or mechanical properties in a controllable manner. By investigating how to improve the controllability and scalability of programmable matter, we aim to unlock its potential for a wide range of applications, from responsive materials to reconfigurable devices.

Self Folding

Self-folding, a self-assembly process where a structure autonomously transforms from a flat sheet into a 3D shape, is a key focus in our exploration of programmable matter. Inspired by origami principles, self-folding offers significant advantages over traditional methods of rapid fabrication for 3D objects, such as reduced material waste and improved cost-efficiency. These structures are particularly well-suited for robotic systems, as they allow for the integration of multiple materials, including actuators and sensors, directly into the sheet. This essentially creates a flexible printed circuit board capable of performing robotic tasks with minimal wiring and manual assembly. The logistical benefits of self-folding systems have garnered significant interest, as these devices can be transported in a flat state and deployed autonomously into complex geometries at scales ranging from millimeters to centimeters.

Re-Programmable Matter

Reprogrammable matter systems open new possibilities for on-demand device customization. The ability to change shape provides a machine with adaptability and robustness and a potential for more efficient task execution. Imagine new computer displays that morph to show objects in full 3D rather than 2D images on a flat screen, resulting in new intuitive modes of virtual interaction. Or, everyday objects that autonomously morph their form: furniture that can serve alternately as chairs, tables, or shelves; smartphones that grow legs and crawl away, then shrink back to fit in a pocket; or cars that change their tire size or treads to suit the terrain.

MORF (Magnetic Origami Reprogramming and Folding System)

To move closer to this reality, our lab has developed an autonomous system for real-time reprogramming of self-assembling, shape-changing, and reusable structures using an origami-inspired approach and magnetic forces. This system controllably folds complex 3D structures through magnetic writing. MORF (Magnetic Origami Reprogramming and Folding System) enables repeatable, fully automated print-and-fold manufacturing of intricate 3D structures, naturally integrating active materials and electromechanical components. By precisely writing a magnetic program onto a laminate sheet at the millimeter scale, the sheet can be programmed and reprogrammed to fold and unfold into multiple shapes at various sizes and resolutions.

Most recently we haven shown how MORF has evolved into a fully reconfigurable platform for robotic and engineering applications by integrating a thin thermoplastic “locking” layer. When heated, this layer softens to allow folding under magnetic control; once cooled, it stiffens to preserve the desired shape with no need for continuous actuation. Through this approach these new structures withstand repeated folding and unfolding cycles and reliably bear loads of more than forty times their own weight. We demonstrate how this can serve as a reprogrammable robotic tool—switching between multiple grip sizes for bolts or screws on the fly—highlighting its potential for adaptive, on-demand manufacturing and assembly tasks.

Zooming out, we expect that magnetic program-and-fold technology will enable new levels of complexity and customization for future reprogrammable matter. Although we have focused in this system on only rigidly foldable patterns where the bending of the sheet is localized at the folds, the same techniques can be applied to applications requiring other types of sheet deformation, such as bending, torsion, or shear.

Heat Based

One method of self-folding involves utilizing a multi-layer structure, where a heat-sensitive contraction film (polyvinyl chloride; PVC) is sandwiched between two rigid structural layers. The self-folding crease patterns feature differentiated gap widths on the front and back faces. Upon heating, the difference in these widths causes the PVC to contract disproportionately, resulting in a bending motion that drives the folding process.

As the complexity of the folding pattern increases, so too does the likelihood of errors due to the multiple potential pathways the folds can take. To address these challenges, we are investigating the self-foldability of non-rigid origami patterns, such as the origami hyperbolic paraboloid (hypar), which is known for its symmetry and bistability.

Recent Publications

MORF: Magnetic Origami Reprogramming and Folding System for Repeatably Reconfigurable Structures with Fold Angle Control

Unger, Gabriel; Shenoy, Sridhar; Li, Tianyu; Figueroa, Nadia; Sung, Cynthia

MORF: Magnetic Origami Reprogramming and Folding System for Repeatably Reconfigurable Structures with Fold Angle Control (Conference)

IEEE International Conference on Robotics and Automation (ICRA), Forthcoming.

(Abstract | BibTeX | Links: )

Re-programmable Matter by Folding: Magnetically Controlled Origami that Self-Folds, Self-Unfolds, and Self-Reconfigures On-Demand

Unger, Gabriel; Sung, Cynthia

Re-programmable Matter by Folding: Magnetically Controlled Origami that Self-Folds, Self-Unfolds, and Self-Reconfigures On-Demand (Conference)

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

(Abstract | BibTeX | Links: )

Increasing Reliability of Self-Folding of the Origami Hypar

Liu, Addison; Johnson, Mykell; Sung, Cynthia

Increasing Reliability of Self-Folding of the Origami Hypar (Journal Article)

In: ASME Journal of Mechanisms and Robotics, vol. 14, no. 6, pp. 061003, 2022.

(Abstract | BibTeX | Links: )

Microfabricated foldable wings for centimeter-scale microflyers

Cha, Wujoon; Campbell, Matthew F.; Popov, George A.; Stanczak, Christopher H.; Estep, Anna K.; Steager, Edward B.; Sung, Cynthia R.; Yim, Mark H.; Bargatin, Igor

Microfabricated foldable wings for centimeter-scale microflyers (Journal Article)

In: Journal of Microelectromechanical Systems, vol. 29, no. 5, pp. 1127-1129, 2020.

(Abstract | BibTeX | Links: )

Self-folded soft robotic structures with controllable joints

Sung, Cynthia; Lin, Rhea; Miyashita, Shuhei; Yim, Sehyuk; Kim, Sangbae; Rus, Daniela

Self-folded soft robotic structures with controllable joints (Conference)

IEEE International Conference on Robotics and Automation (ICRA), 2017.

(Abstract | BibTeX | Links: )

Miyashita, Shuhei; Guitron, Steven; Ludersdorfer, Marvin; Sung, Cynthia; Rus, Daniela

An untethered miniature origami robot that self-folds, walks, swims, and degrades (Conference)

IEEE International Conference on Robotics and Automation (ICRA), IEEE, 2015.

(Abstract | BibTeX | Links: )

Miyashita, Shuhei; DiDio, Isabello; Ananthabhotla, Ishwarya; An, Byoungkwon; Sung, Cynthia; Arabagi, Slava; Rus, Daniela

Folding angle regulation by curved crease design for self-assembling origami propellers (Journal Article)

In: ASME Journal of Mechanisms and Robotics, vol. 7, no. 2, pp. 021013, 2015.

(Abstract | BibTeX | Links: )

Current Personnel

  • Gabriel Unger (MEAM PhD)

Acknowledgements

This work was supported in part by the Army Research Office (ARO) under MURI Award #W911NF1810327, by NSF Grant1845339, NSF Grant No. 1659190, ONR Award #N00014-23-1-2068 and the Johnson & Johnson WiSTEM2D program, by the Penn Health-Tech program, and by the Penn Center for Undergraduate Research and Fellowships. Any opinions, findings, conclusions, or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the funding source.