8. Soft robotics¶
Research & Inspiration¶
This week we explored soft robotics and soft actuators. Everything I made this week is an actuator because they are not yet connected an an input or designed to produce an output.
We traditionally think of robots being made of metal, plastic, or other rigid materials, even in their joints. Soft robots are systems that are compliant and flexible. They possess sensing capabilites and can receive input from their environment. They are robots which are made with soft materials.
In the practice of soft robotics, inventive principles in nature are translated into engineering solutions. Bionicists connect structure to function.
- Saskia Helinska, Unflatables Fabricademy Final Project
- Anicka Yi, In love with the world
- Toyota Research Institute, Punyo (gripper tech is open source)
- David Blow (article author) Face taping for Bell’s palsy
- James Merry for Bjork
- Jun Murakoshi, Noisy Instrument
- Polina Osipova, Cyber Svetlana
- Eunyoung Park, Radical Soft Robots
I broke my research and search for inspiration down into two categories: funtion and aesthetic.
During my time at Toyota, I learned about the mobility initiatives of the company. One such project is Punyo, a robot designed to aid in tasks difficult for mobility impaired individuals, like lifting and moving large items. The technology for Punyo's "Bubble Gripper" soft robotic paws has been open-sourced by the Toyota Research Insitute. I noticed that many of the applications and projects we saw this week are being done as academic research, and wanted to share something that is being applied and created outside of an academic intitiution.
weekly assignment
Check out the weekly assignment here
References & Inspiration¶
Fabricademy Participants
-
Inflatables- Maricruz Garcia
-
Gelatin- Batoul Omar Al-Rashdan
-
Wearable- Montse Ciges
I found the functions of my interest leaned towards wearable devices, and plan to further develop my ideas started here during wearables and skin electronics weeks!
I am referencing a personal medical experience in this project. I looked into research on treatments, both proven and anecdotal for the development of my soft robotic wearable.
In 2024, I had Bell’s palsy for three months. It’s a condition that causes paralysis/weakness of the muscles on one side of the face. Inflammation of the seventh cranial nerve cuts off signals between the brain and the facial muscles. It can be painful because inactive muscles tighten and shrink. Acupuncture, facial massage, and applying gentle heat to the area were all methods I used to manage Bell’s Palsy. 13 months from the onset of it, I still have some long-term effects of the nerve damage and incorrect nerve regeneration.
"Biofeedback is a technique that enhances sensory feedback, enabling individuals to consciously modify bodily functions typically considered involuntary. This technique is commonly used for symptom management in chronic illnesses and as part of physical therapy for patients with motor dysfunction." (NIH)
Tools¶
Tools
- Arduino UNO
- Arduino IDE
- Heat press
- Vinyl cutter
- Heat-press vinyl and TPU sheets
- Baking paper
- Cutting mats and X-acto/box cutters/scissors
- Laser cutter
- Air pump and tubing
- 3D modelling software (Rhino)
- 3D printer and slicer software
- Flexible silicone (Smooth On EcoFlex)
- Gelatin, Glycerin, Sodium Alginate, Calcium Chloride for biomaterial experiments
- TPU and PLA filaments
- Hydrogel beads
Process and workflow¶
Our goal this week was to explore different processes as a group to gain a better understanding of this concept. Nobody really had previous experience in (soft) robotics.
Heat press¶
We followed this technique shared by Adriana Cabrera during this week's lecture.
I started by experimenting with the heat press material: vinyl and TPU. I also wanted to understand the mechanics of bending with these designs. I struggled to get the vinyl to fuse in the hinges I made. I think some of them were too thin, but the ones that worked didn’t have enough variation in their dimensions to bend in one direction.
The TPU worked better, and the baking paper inside provided a rigidity that allowed them to bend/curve in one direction. I removed the baking paper from inside the TPU pieces with a crochet hook and determination, and they lost the ability to curve or bend.
Cast materials¶
After struggling to successfully apply the mechanics of bending in my pneumatic devices, I began looking into creating pressure for sensory input instead. The diagrams Saskia Helinska designed to explain the mechanics of her silicone inflatables became a starting point for a wearable device to apply gentle pressure to areas of my face.
I used Rhino to make 3D printable molds for casting silicone. I based the model on the shapes and mechanics of my successful TPU inflatables.
I drew air channel shapes flat, used offset to make the outer edge and walls of the mold, and then extruded the curves to make the drawing into a 3D object.
I am still testing this design. I had a feeling that mine would not inflate evenly, after seeing Asli blow into the tests we did with a molds from previous years.
Ironically, my face is not strong enough to manually blow air into my piece. I designed another, more finalized wearable piece to try out after testing my temple patches. Maybe I will incorporate manually inflating it into my wearable design as a means of physical therapy for my lip and cheek muscles.
Unfortunately, my attempt to 3D print this got messed up. I am working on optimizing the file to try again this week. I may have to print it at a slower speed than the generic PLA preset gives. I may also see if changing the orientation of the design on the print plate helps.
My goals going forward:
- Find a happy medium for the air pressure inside the chambers. The bike pump is too strong, but manually I am not strong enough to test it. Our arduino-controlled pump/vaccuum broke, but the air pressure was able to be manipulated.
- Get a successful 3D print for the wearble mold. Maybe it needs to be TPU/waterproof polyester/vinyl with lasercut paper inside.
- Test the wearable design. In my design, the air chambers are the same size, but there are longer air channels. See if it will inflate evenly.
- Program inflating and deflating with Arduino. Get tubing, connectors, and decide how to attach it to my face.
Group experiments I am not personally exploring further¶
Gelatin¶
I wanted to test pneumatic actuation with gelatin, but didn't have high expectations for getting it to work. In the past I tried another project where I attempted to fuse gelatin bioplastic pieces together but it didn't work. Asli suggested the process, but after doing it I came across Batoul's documentation which is more technical than our attempt, so I would follow that if I tried again.
I made the bio-silicone, then Flora, Asli, and I tried to seal the pieces together. The clover-shaped air pocket was the only one that inflated a bit but all pieces has problems with coming apart.
We used pre-existing molds, and a higher glycerine/gelatin:water ratio to have less shrinkage.
Gelatin recipe
- 80 g gelatin + 80 g glycerine + 300 ml water
- Heat water, add ingredients, and dissolve. Stir gently and simmer for 20 minutes. Cast sheets and mold pieces, and let them dry.
- Melt leftover gelatin with a small amount of water. Pour in areas you want to seal and quickly attach the top and bottom pieces.
- Allow to cure/dry before attempting to inflate.
Hydrogel actuators¶
Based on AquaMorph project presented at Dutch Design Week by Yaya Huang. We didn't see it this year but Asli saw this the at a prior year's expo.
We made an actuator from 3D printed TPU filled with hydrogel balls. When soaked in water, the expansion of the hydrogels inside the cavities causes the arms to bend and twist. Asli conducted this process, and we documented it together.
In the time it takes to pause the print and fill the cavities with the beads, the TPU cools too much and doesn’t fuse well after starting the print again. This caused some of the cavities to burst open from the pressure. We tried to prevent this by applying diffused heat from a heat gun while filling the cavities, but it didn’t really help. Asli also tried fusing it post-printing with waste filament and a soldering iron, but that wasn’t working either. Despite some of the design issues, the overall actuation was still achieved.
Sodium Alginate¶
We followed the technique from Hala Amer’s video tutorial, but with a different recipe and process for making the alginate gel.
We had leftover alginate from biomaterials week and wanted to do something with it. We ended up having to use zip ties to secure them to the tubing.
Alginate Gel recipe
- 12 g alginate + 40 g glycerine + 400 ml water
- Blend water, alginate, and glycerine with immersion blender or whisk until no clumps remain.
- Allow the mix to sit overnight to release trapped air bubbles.
- Make a calcium chloride solution (two heaping tablespoons dry calcium chloride pearls/L water , or 1:2 ratio liquid calcium chloride:water)
Alginate Gel bubble process
- Place a thick woven fabric like denim in a wooden frame and soak it with calcium chloride solution, including the sides of the frame.
- Pour alginate gel in a thick layer, ensuring it reaches the sides of the frame.
- Spray the top of the gel with calcium chloride solution and allow it to cure for 30 seconds or until it begins to shrink away from the sides.
- Poke a hole in the gel skin, slightly above the side seam, and insert tubing.
- Apply pressure to the gel bubble to move some gel to surroung the tubing, and spray with calcium chloride solution to cure/secure the tube to the bubble.
- Inflate the bubble with air, then use a syringe to push the calcium chloride solution inside the bubble. Continue this process
Also, the one I helped make was cured for too long, and the skin became tough to pierce and put tubing inside. Tajah and Amber kept going with this process, so check out their documentation.
3D Models¶
find .3mf below1
find .3mf below2