Project Objectives

To create a tactile experience that captures the nuance of human touch through gesture and movement. The wearable bracelet presents a novel approach to appreciating the complexities of human interaction and enhances our ability to communicate and connect with one another on a deeper level.

Reflection on the creative design opportunities

In our world, human gestures and phrases carry nuanced connotations not captured by verbal or visual communication. Soft robotics provides a unique opportunity to explore the subtleties of non-verbal communication and how they can fill in the gaps. In this project, we explore the use of silicone to create gesture-activated bracelets that aim to mimic the tactile experience of human touch.

Throughout the design process, we embrace the skin-like qualities of silicone and the aesthetic material changes during pneumatic actuation to communicate non-verbal affection and comfort in response to a gesture of greeting. The design of the bubbles on the bracelet and how they felt and looked on the skin was key consideration. The process of perfecting these involved experimenting with different shapes, sizes, and textures. Choreographing the actuation of the bracelet involved experimenting with the speed and pressure of air to create pulses that represent what the sender wants the receiver to feel when they perform a gesture. This required a deep understanding of the subtleties of each gesture and how they can be translated into a physical experience.

The marbling and coloring of the silicone offer an artistic interpretation of the grotesque nature of human flesh. This serves as a representation of the raw emotion, intimacy, and vulnerability that can be associated with touch-based communication.

Outcomes

SUCCESSES

Overall, we are very happy that we achieved creating an actuating bracelet that had such an interesting feeling and look to it. There are always more opportunities and avenues we wish we had time to further explore, but we did experiment a lot throughout, so we feel satisfied with our iterations. The downfall to lots of experimenting is that we spent a lot of time failing instead of moving toward a final form. We mainly played with different ways of actuating (phasing, alternating, and a gradual wave-like actuation), size, thickness, length, color, and sensors. However, our vast iterations proves we were willing to try things that were challenging and still found many great additions for our final bracelet. For example:

• We figured out early on that we enjoyed the bubble form and how interesting it looks and feel. We were able to easily iterate to make the bubbles expand properly on both sides. This created a nice sensation on the skin while wearing the bracelet. This was a consistent success with our prototypes and the final form.
• We created a mold that successfully caused an alternating actuation, which was incredibly cool, but wasn’t included in the final iteration because of other challenges and lack of time.
• We liked almost every color we made and got to have a lot of fun experimenting with this in the end. We believe that our colors create a very interesting story in the final product that really helped us create our story into something compelling
• Tying the bracelet together was interesting because it was less iterated on, but we found the sewing to be the easiest and most secure way, which actually caused us to create some interesting bracelets. We sewed a bracelet together which ended up looking really cool and it was a great iteration in our collection. Overall, the sewing was quite secure and we didn’t have to worry about it ripping or falling off.
• The accelerometer sensor ended up being very cool and we were able to make 3 different gestures and choreographed 3 different actuations for each gesture by changing the input to the air pump. We were also able to actuate the two sides of the bracelet with different timing, which made the actuation look more complex and interesting, kinda like a wave. This was a great success to us because we think it makes the interaction with the bracelet more interesting by creating multiple ways to communicate through this bracelet rather than having one input and output.

CHALLENGES

1. Bonding the silicone. Our bracelet mold had thin tubes running through it for actuation purposes which caused a lot of issues during the bonding process. The tubes get getting plugged which resulted in the actuation/inflation of the bubbles being very uneven.
2. 3D Printing. The 3D printers offered were not long enough to accommodate the full length of the bracelet which meant that we had to figure out a way to connect two silicone parts seamlessly. This proved to be pretty difficult and resulted in a very fragile bracelet. The IDEATE printers also broke halfway through the course so we couldn’t experiment with different connecting part designs that may have resolved the issue above.
3. Suctioning air out in the motor + pump system. When we created the system to actuate the bracelets we found that they didn’t vacuum the air out which meant the bracelets were at risk of popping (since the top layer was very thin). To work around this, we poked holes in the tubes to allow the air to leak out when the bubbles deflated. This allowed us to control the inflation and deflation in a way that was very natural.
4. Gesture detection vs classification. Initially, our plan for gesture detection was to collect data with the accelerometer to train a classifier for predicting gestures. However, there wasn’t enough memory on our Arduino and filtering out data points was outside the scope of the project so we ended up going with a gesture detection system based on thresholds of x, y, and z values on the accelerometer.

Citations

Technical documentation

Source code:

Original CAD files:

Photographs:

https://drive.google.com/drive/folders/1otIVnCnxDc3Mx3z4TsxX-oGGNbQXWG-d?usp=sharing

Video:

Contributions

Gia

• wired up circuitry on the breadboard and accelerometer
• Sewed bracelets together
• made a box for final video
• edited final photos and video
• modeled bracelet for video and pictures

Fiona

• created the CAD for the mold used in the final product
• casted, bonded, and colored silicone bracelets for the final demo and video
• choreographed the actuation for the final bracelet gestures
• coded a state machine based program for gesture detection using an accelerometer
• took pictures and video for final documentation

Both

• ideation
• created CAD molds and casted and bonded silicon prototypes
• tested prototypes and documented this
• sent in 3D prints and acquired them
• wrote part of documentation
• tested heart rate monitor in class
• made velcro for accelerometer

gesture detection

We first found the thresholds for the four directions (left, right, forward, back) through trial and error and created conditions in the main loop to check for changes in the x, y, and z values that crossed these thresholds. We took inspiration for the algorithm from this paper: https://www.irjmets.com/uploadedfiles/paper/volume3/issue_2_february_2021/6052/1628083249.pdf

Then, we took at the axes of movement for three gestures: normal wave (left to right), cat wave (up down), and a palm flip. Based on the direction of movement across a time window, we were able to create loops in the main sequence to detect these movements.

Overall, this week we made good progress. We were able to 3D print our mold and cast our part for the final bracelet. We were also able to create a program that detects 3 types of gestures.

Casted Bracelets

https://drive.google.com/file/d/19sY36DcLptIkVzm0aVcRVbvsjIRpSGyS/view?usp=sharing

(our video testing our prototypes)

We casted a bunch of prototypes and we are trying to put together our final bracelet. We tested them all today and we are running into some issues with the tubes keep popping off, but we think we got some finalized versions that should work once we tie it together in a bracelet!

Here is us testing the whole bracelet mechanism together with all the gestures!

Final test prototype to see if it will work to silicon two bracelet parts together:

New objectives for the upcoming week

• Putting everything together: bracelet construction and making box with tube connection and making sure everything is seamless for show and documentation

need to buy

• idk maybe string if we want nicer string?

End of Week Report

We finally got our molds 3D printed so we have casted the silicon into our molds. We still need to test the silicon pieces with an air pump. We have started the circuit for connecting the air pumps and Arduino, so once we know that is working it will be easy to connect the accelerometer. Lastly we have thought about how to connect the bracelets together and be put on the wrist. We are thinking some nice looking string for the final design, but we need to test that with our need silicon prototypes.

New objectives for the upcoming week

• cast and bind the new molds
• decide if we need to 3D print any other molds
• send final 3D prints
• finalize design for gestures to detect
• figure out accelerometer code

need to buy

• idk maybe string if we want nicer string?

End of Week Report

This week has been mostly testing and fixing our CAD models and trying to get them printed. We realized that our molds needed some slight changes so that they would print properly, it mostly had to do with changing the ports and tubing to fit correctly. We did a lot of testing with our silicon molds and how it is going on the wrist, mostly seeing how the bending affects the airflow. We are trying to figure out how it will exactly go on the wrist, we tested velcro and thought about how string or silicon would be incorporated. We also tested how the heart rate monitor would go on the wrist and how much pressure it needs. It was very informative and helped us start imagining exactly how we will attach it to the wrist. We think velcro is a very viable option.

New objectives for the upcoming week

• start to figure out final design
• silicon cast our 3D prints
• finalize how we are attaching bracelets

To buy: nothing

new mold 1 – full bracelet prototype

This is a mold for a full bracelet with the ability to be tied together or clasped? If you fill this whole mold up it should make one full bracelet. Also this has the thinnest tube we have done so far with a .5mm hole size.

new mold 2 – 2 tube system

We also made a mold that’s gonna test how it would actuate if we had two air pumps, so this caused us to make a much different mold with 2 tubes running along the sides and inputting to every-other cylinder. This is our last actuation experiment before refining what we have already tested.

Last weeks prototyped glued and tubed

New objectives for the upcoming week

• test new heart rate monitors if they are available
• get molds printed and casted (if skylab allows us)
• decide what actuation style we have liked so far and go forward with that. Figure if it needs refining. Definitely should start CADing another mold before this week ends

To buy: nothing

• Phasing/ sequencing
• Using 2 pumps for one bracelet
• Using modularity to make a full bracelet
• against skin actuation

Beginning of the week plans

We want to print 2 different molds this week to test the concepts above. This is a rough first plan before attempting anything.

Things we need to figure out in the future

• How will it communicate heart rate 
• Where are the sensors gonna go, heart rate monitor 
• feedback and feel?
• connecting pumps and where/ how to place the electronics and pump on the body

End of Week Report

For this week we got two molds CADed to test out two different variations of actuation. In one mold we experimented with the inner tube being much narrower and in the other we made the inner tube taper into a smaller diameter. We are hoping to see a phasing out of these changes. One of the prints failed which we found out on Friday, so we re-sent it to the printer but we didn’t get to fill that mold sadly. We did start a 3rd mold to continue progress!

We attached some photos of the CAD for the mold below:

New objectives for the upcoming week

• print a 3rd mold and test a new actuation method
• make a full bracelet prototype

To buy: We think we need the other heart rate monitor because the one we got is pretty fickle

What would a machine mind dream of after “seeing” the vast collection of The Museum of Modern Art? In other words, if the corpus of images of the MoMA collection had been accomplished by a single artist, what would their dreams look like?

STATEMENT OF ARTIST INTENT

Unsupervised is a meditation on technology, creativity, and modern art. Anadol trained a sophisticated machine-learning model to interpret the publicly available data of MoMA’s collection. As the model “walks” through its conception of this vast range of works, it reimagines the history of modern art and dreams about what might have been—and what might be to come. In turn, Anadol incorporates site-specific input from the environment of the Museum’s Gund Lobby—changes in light, movement, acoustics, and the weather outside—to affect the continuously shifting imagery and sound.

ALTERNATE EXPRESSIONS WITH SOFT TECHNOLOGIES

Fluid dynamics has always been a massive source of inspiration for Anadol’s “Machine Hallucinations” project. The artist’s exploration of digital pigmentation and light through fluid solver algorithms could also inform the kinetic effects of soft sculptures that mimic fluid movement. Furthermore, his exploration of algorithms to simulate the movement of fluid could inspire the work of other AI-based artists.

APPLICATIONS TO THE ART CONTEXT: in what way will the technological expression be artistically meaningful?

The artist’s vision is to create a piece of work that can “handle data within a universe that it creates for itself” and strives to do this with his approach to data visualization’s latent space “as a locus for never-ending, self-generating contemplation”. The way Anadol brings together rich threads of information to weave beautiful visuals that incite feelings from viewers shows how data and generative algorithms can be used to create artistically meaningful work.

TECHNICAL PAPER WHICH ADDRESSES RELEVANT TECHNOLOGIES

Fluid Mechanics of Pneumatic Soft Robots

• this research investigates the movement of gas within an elastic field and explores how it can be applied to various realistic actuator configuration
• it provides a foundation for modeling fluid dynamics within fluid-actuated soft robots

Bio-Inspired Wearable Robotics for Fashion Technology: SoftVoss

This was an exhibit created by Dr. Yin Yu for the 2022 San Diego Design Week. In the video Dr. Yin Yu shares how to design a bird-feather-inspired wearable fashion technology with soft material, highlighting why biomimicry design in soft robotics creates engaging human-computer interaction empowering designers’ creativity.