53399 Creative Soft Robotics

Project Objectives

The objective of this project was to utilize soft materials and a material class called “auxetics” to demonstrate the ways that shape and form can create unexpected actions and reactions. These metamaterials, so called due to their properties depending on their shape as well as the material they are made of, have a negative poissons ratio, which means they expand in two dimensions under a single axis load. Compared to traditional materials, this creates an unexpected and visually distinct effect that gets the viewer thinking more about shape and motion.

Creative Design Opportunities

Fabricating our auxetic designs using silicone materials allowed us to test the designs from established literature and modify them. The iterative process enabled us to test these structures at different scales, thicknesses, and densities to create the most dynamic and interesting patterns when applying tension. To avoid unnecessary work, we used the built-in stress simulation in Solidworks to make educated predictions on how the material will behave. The silicone fabrication process verified the relative accuracy of the simulation, which reduced our iterative workload significantly.

Not only does the fabrication process validate our assumptions, but it also allows us to modify and experiment with the current iteration. Using Rhino Grasshopper, we transformed and wrapped the grid-like design. However, the simulation was unable to manage the complexity of the form. Therefore, we had to cast and test the silicone with our hands.

Overall, the soft technologies and the fabrication process allowed us to iterate, improve, and realize our designs.

Process

Research

Our first step was to find past designs that had been proven to work. A deep dive into literature allowed us to find 4 main categories that we ended up modeling and testing. Each of these had distinct visual and motion properties, which we were only able to find out by running FEAs on the designs that we made. Further research allowed us to refine our designs, both by reviewing literature and by validating new designs. This allowed up to pare down the options to a few final designs that we would end up presenting. Our aim was to find unique angles on these auxetic designs by changing parameters, such as taking away links or distorting the mesh. Our continued testing process allowed us to create the most interesting effects on our final designs.

Digital

We used Rhino to create some of the existing auxetic designs from the previous research literature to design the 3D-print mold. This process was relatively quick and straightforward; however, we encountered the most difficulty transforming a design in Grasshopper.

Garth gave us the idea of adjusting an existing design, usually in a grid-like pattern, to see the outcome of such a transformation. This suggestion led to us developing a complex parametric Grasshopper script that uses attractor points and parameter controls. We could modify an existing pattern into two new patterns: expansion and contraction. However, due to issues with the scale and sizing of components, the contraction iteration could not be displayed at the final critique on April 26th.

Physical

As previously stated in Creative Design Process, casting silicone allowed us to verify our simulation results and test for new designs, but it also helped us to fine-tune the thickness and density of our designs. In our earliest iteration, we used one of the softer mixes of silicone, which resulted in the mesh failing structurally and breaking. Learning from our poor choice of materials, we opted for the stronger mix of silicone, which worked out great in the end. The stronger mix allowed us to apply more stress pressure on our mesh and realize its full structural potential.

Outcomes

The final three designs we chose were a uniform auxetic, a mesh of hexagons and auxetic combined, and one curved auxetic that was warped using an attractor point. These provided an interesting sequence of behaviors: the uniform introduces what an auxetic is, the combined contrasts that with an expected and known material, and the warp provides distinct and new interaction.

To house these meshes, we designed frames that served three main purposes. The first was to provide a solid way for the viewer to interact with the pieces. The frames had handles on them that attached to the meshes, allowing for distortion in a controlled, planned manner. The second purpose was to allow for a reference point that would show how much the mesh distorts. Being housed in a constant area while the form expands draws attention to how the shape of the mesh changes. The final purpose of the frame was to provide a way to lift the mesh off the table, which allowed for lighting effects. Since these meshes are very distinct in their shapes, we wanted to highlight how they interacted with light. We decided that shadow effects would be most interesting, and set up two above spotlights to create unique shadow effects under the frames.

Throughout the process, we learned that we should have tested our models physically more, rather than relying on the FEA for the designs. Having more physical models would have allowed us to develop more interesting shapes and behaviors. We also learned to focus more on the true purpose of the project. During each step, we tended to lose sight of the big picture, and instead focused on applications. We were guided to be more pragmatic and think about how each decision impacted the artistry of the piece, rather than trying to rush for an interesting application.

However, this project was certainly a success by our own standards. We were able to create really interesting and unexpected behaviors from our auxetic meshes. We succeeded in our purpose of making surprising and pleasing deformation, and the hands on aspect of the final designs helped in this effect. The viewers during the final presentation were really taken by surprise with the behavior of these meshes, which is exactly what we hoped for.

Video & Images

Citations & Supplemental Materials

Mizzi, Luke, et al. “Auxetic Metamaterials Exhibiting Giant Negative Poisson’s Ratios.” Physica Status Solidi (RRL) – Rapid Research Letters, vol. 9, no. 7, 2015, pp. 425–430., https://doi.org/10.1002/pssr.201510178.

Ren, Xin, et al. “Auxetic Metamaterials and Structures: A Review.” Smart Materials and Structures, vol. 27, no. 2, 2018, p. 023001., https://doi.org/10.1088/1361-665x/aaa61c.

Pan, Qi, et al. “Programmable Soft Bending Actuators with Auxetic Metamaterials.” Science China Technological Sciences, vol. 63, no. 12, 2020, pp. 2518–2526., https://doi.org/10.1007/s11431-020-1741-2.

Grasshopper Script:

https://drive.google.com/file/d/17Q0tD4kFrwGaWCBIePoLLMq2cEud6FP1/view?usp=sharing

Final Product & Frame Designs:

https://drive.google.com/file/d/1mG3Nu2QUIlcHzCPTOhHCXspGPzkfqKoY/view?usp=sharing

Contributions

Dunn: 3D printing molds, final mesh casting, frame design, frame cutting and assembly, Rhino modelling, attractor point setup

Elliot: Literature review, design selection, auxetic casting, frame cutting and assembly, FEA simulations, lighting and presentation setup, model cad for some meshes

The images above are our first 3D-printed test cases, printed with soft material at different thicknesses. The one on the left is thicker, and the one on the right is thinner, without any other differences. We wanted to see the capacity for these structure systems to expand when tension is applied in the four major axes. They do produce an expanding form change. However, the extent is lacking, especially with the thicker material. Though much stretchier, the print on the right is fragile.

In our latest experiment, which result is yet to be studied, we used the same pattern at a denser scale and with silicone instead of 3D printing. We hope this softer material would have greater flexibility; however, they may lack the rigidity the 3D-printed material has. Depending on how successful this test is, we may or may not revisit and readjust our concept.

Our concept is to create a soft-actuation-driven dynamic architectural facade element that enhances building performance. The American Institute of Architects (AIA) has established a goal for 2030: all proposed and built projects must be net-zero. Whether or not that is achievable is not within the scope of this research.

Our system uses a soft, flexible, and robust material as a mechanical pump, creating a negative pressure in a closed-loop heating system. High-performing architectural facades inspired this proposal in locations farther from the equator, which sees relatively higher temperatures during daylight hours and vice versa. A few architectural terms must be addressed to understand this project proposal fully.

• Low-E Coating
  • A type of metallic, invisible to human eye coating that is applied onto glazing assembly to reduce infrared light transmission rate significantly. It is often used to allow natural light to enter and block heat.
• Glare
  • Direct light creates a highly uncomfortable indoor environment.
• Internal Heat Gain
  • Somewhat related to the purpose of Low-E Coating. Internal Heat Gain refers to the increasing temperature inside a building, usually caused by solar radiation penetrating the glazing assembly. Still, it can also penetrate via opaque surfaces much slower.
• Lux
  • A light measurement unit.

The plan above depicts the two states of our system. Upon exposure to daylight or heat, the soft pump will activate and enlarge, pulling water resources from an off-site location. Then, the water inside the pump structure would be maintained inside and gradually heated up due to solar exposure. To accelerate this process, Low-E coatings are applied to the interior of the assembly chamber, effectively trapping heat. This process will also reduce the building’s internal heat gain. At the same time, the enlarged pump structure will serve as a shading device, reducing glare and creating indirect light.

Once the exterior temperature or lux drops to a certain threshold, the heated water will be released inside a closed-loop heating system via tubes connected to the pumps. The heated water can then be used as hot water for lavatories or as floor heating using a pex-tubing system.

We are currently investigating the affordance of the soft pumping mechanism, especially the volume differences between the two states. Ideally, the greater the difference, the more efficient and cost-effective this system will be. However, other issues regarding the system invites future studies, such as maintenance, material, and thermal-exchange performance.

This prototype is unrelated to the project Elliot, and I are working on. Instead, an experiment in learning the mold design process and the silicon cast workflow. In the final product, there will be a chamber inside that gets wider in the middle and skinnier towards the two ends. And, in theory, when pumped with air or liquid, the middle section would expand more than the ends.

Abu Dhabi (Completed June 2012)

The Al Bahar Towers’ dynamic facade design drastically reduces the building’s direct solar exposure, internal heat gain, energy expenditure, and elemental damages such as sand erosion. The facade is an external curtain wall, offset from the glazing by two meters. The design of the panels is inspired by ‘mashrabiya,’ a traditional Islamic lattice shading device. At night, all the panels will open(right), and during the day, the panels will close as they follow the sun’s trajectory.

Abu Dhabi’s climate is an arid desert with only two main seasons – winter and summer- separated by two transition periods. Summer is undoubtedly the more extreme season, seeing average daytime temperatures rise to around 45C, with humidity levels at 80% to 90%. At night, the temperature drops no lower than 30C. The Al Bahar Towers’ dynamic facade currently operates on pre-calculated sun patterns and meteorological data, given that the regional climate is extreme yet stable. The same intent can be achieved by a light-sensitive or temperature-sensitive, phase-change material that responds to the change in light exposure or temperature fluctuation—making the building more adaptable to its climate and eliminating unpredictability. The design can still be in the configuration of panels, as it would be cheap and easy to produce. I envision a ballon type of material/construction of the panels, where individual ballon will inflate/deflate based on temperature or/and light to open and close.

Flexural biomimetic responsive building façade using a hybrid soft robot
actuator and fabric membrane

https://reader.elsevier.com/reader/sd/pii/S0926580522005301?token=583A92DCA46D35C9E00BCD8C78D84DF34D91E38244D90EF78E30F29EAF1B042F359926188A037551C18128B67613FDC9&originRegion=us-east-1&originCreation=20230206035857

As the understanding of building technologies and material sciences advances, architecture also becomes amorphous and self-shaping to adapt to the ever-changing environment. The paper by Mi-jin Kim, Baek-Hyeon Kim, Je-sung Koh, and Hwang Yi investigate the affordance of bio-inspired and nature-inspired kinetic responsive facades (KRF) in architecture via an experimental study regarding soft robots. The experimental system uses a combination of rigid frames, shape memory alloy springs, and soft actuators to drive locomotion by air pressure. These operations can be controlled manually or by sensing the environment, such as light and temperature.

The facade design proposed in the research paper can be applied to the Al Bahar Tower and many other pieces of architecture. The unique form will provide visual stimulation and real-world energy-saving potentials. Combined with automated control systems, the entirety of the building facade can become a self-regulating entity that responds to its environment.

This is actually the project of my current architecture professor, the former head of the School of Architecture, Steve Lee’s project. It was done when he was a sophomore at CMU (1979), alongside other students; some became my other professors. This project investigates the affordance of using robust and elastic membrane materials to construct scaffolding, and then utilizing extreme cold to freeze water between two layers of membranes. It was intended to be an architecture sculpture meant to be built during winter and melt once spring arrives.

The process is documented in the bottom images. Two layers of membrane serve as the container for the water, which volume is calculated based on intended building size, support strength (bottom top left, skinny metal support elements), maximum deliverable air pressure (used as a way to pressurize the interior, therefore creating an invisible supporting element), and dome curvature. These calculations and plans allow for a very thin, large, and strong structure. And the construction process, though solely dependent on the weather, can be completed overnight.

This process uses an elastomeric material as the scaffolding for liquid solidification, a process that can be replicated in many other applications, such as medical use. Soft, elastomer materials such as this can form temporary structures that can stop bleeding or open arteries.

Some could say this is a scaled-up version of a soft, kinetic actuator, as its motion is controlled by air pressure and built-in membrane strength. The same principles can be applied at a different scales,s for example, a much smaller iteration can use hydraulic instead of air to achieve a more significant internal pressure, therefore, a more robust scaffolding.

DOI: 10.1109/HSI47298.2019.8942609

1. Do you have any conflict of interest in reviewing this paper?

No 

• Expertise. Provide your expertise in the topic area of this paper.

No Knowledge 

• Summary. Please summarize what you believe are the paper’s main contributions to the field of soft robotics.

Though not related to soft robotics on the surface. This paper is the foundation for more human-soft-robots collaboration and interaction. The main concern of human-robot collaboration is safety, as typical robots are made from hard materials, which can easily cause major damage to the human body. Soft, complaint materials, combined with the finding of this paper, could bring about a more effective and safer iteration of human-robot collaboration.  

• Strengths and Weaknesses. What are the main strengths and weaknesses of this work? Does the paper have strengths in originality and novelty?

The paper is strong because it is very original and novel. The concept is untested and new. Therefore, it is also the reason why it has some jarring weaknesses.

Firstly, the paper can be repetitive sometimes. For example, the phrase “dynamic affect-based human-robot collaborative strategy” or phrases alike are often seen in the same paragraph, making the paper wordier and harder to read.

Secondly, the section regarding the process of the experiments was hard to read. The authors’ naming convention for objects, locations, and participants (letters or letters with digits) complicates the reading. I constantly go back and forth on different pages to refer to the figures. Additionally, labeling locations, objects, and participants in the images was not precise.

• Soundness. Are the ideas, algorithms, results or studies technologically/methodologically sound?

Despite my criticisms of the author’s writing style, I find the process logical. The authors first studied the affective features in a human-human collaborative environment and then applied the result in similar human-robot collaborative applications.

One thing I do question is the robot’s expression. I do think the expressions so far are lacking in variety since humans are much more complex emotional beings. However, I do understand the scope of this paper does not encompass psychology. A more complex display of the expression could potentially enhance the collaborative performance.    

• Related Work. Does the paper adequately describe related and prior work?

The paper describes the related and prior work both adequately and succinctly.  

• Presentation. Is the paper well organized, well written and clearly presented?

Overall, the paper is well-written and clear. Some parts can be improved, such as parts mentioned above.

• Suggestions. Do you have suggestions for improving this paper?

Reduce repetition.

Better image documentation

Use a less boring, more engaging way to describe the process. 

• Comments to Committee (Hidden from authors). Does the paper have enough originality and importance to merit publication? Is the paper relevant to the field? These comments will NOT be sent to the authors:

Yes, I believe so. It is not extremely relevant yet to soft robotics; however, I see great potential in the field, especially combining the affect-based robot motion control and soft, compliant material construction, which would make the collaborative process safer, thus, higher performance.

1. Overall Rating. Provide your overall rating of the paper (5 is best)
   1. 5 – Definite accept: I would argue strongly for accepting this paper.

From there, I used Web of Science to access all the papers that cited the root paper. The root paper had over 500 citations.

The three result paper I chose were “4D printing Light-Driven soft actuators based on Liquid-Vapor phase transition composites with inherent sensing capability” (DOI: 10.1016/j.cej.2022.140271), “A Wire-driven Elastic Robotic Fish and its Design and CPG-Based Control” (DOI: 10.1007/s10846-022-01797-9), and “Bioinspired Multi-material Polyjet-printed Frog Robot for Synchronous and Asynchronous Swimming” (DOI: 10.1007/s42235-022-00321-x)

Venous Materials: Towards Interactive Fluidic Mechanisms

Synopsis:

Venous Materials are a novel concept of an interactive material using fluidic mechanisms that responds to deformations generated by user.

Hila Mor, Tianyu Yu, Ken Nakagaki, Benjamin Harvey Miller, Yichen Jia, and Hiroshi Ishii. Venous Materials: Towards Interactive Fluidic Mechanisms. In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems, 1–14. Honolulu HI USA, April 2020. ACM. doi:10.1145/3313831.3376129.

Paper 2:

Soft Robotics: A Bioinspired Evolution in Robotics

Synopsis:

Give soft robots physical capabilities found in animals, especially those that lack an indo-skeleton such as cephalopods, jellyfish, and earthworms.

Kim, S., Laschi, C. & Trimmer, B. Soft robotics: a bioinspired evolution in robotics. Trends Biotechnol. 31, 287–294 (2013). doi: S0167779913000632

Paper 3:

Development of Magnet Connection of Modular Units for Soft Robotics

Synopsis:

A discussion about using the principles of modular design and magnetic joint connections to give soft robots more practical application opportunities.

Jun-Young Lee, Kyu-Jin Cho. Development of Magnet Connection of Modular Units for Soft Robotic. doi: 10.1109/URAI.2017.7992886