Today we worked with our own DIY devices rather than the ‘pre-fab’ sensors. We spent a good portion of the time workshopping the shirt with the actuators and then moved on the other soft circuit garments.
We covered a variety of activities – live coding the buzzers, buzzers triggered by a handmade sensor, and using data from the previous day’s work with the Polar and OmSignal shirt to create a buzzing pattern. In general, the vibe motors that are shown on each sleeve.
The live coding of the actuators worked really well. I created simple keypresses in Processing to send OSC to the xOSC micro controller and turn the buzzers on and off. We tried various scores such as switching between right and left sides or limbs, or stopping moving when sent a short buzz. The dancers really did respond as an output of the coding.
We also explored using the crocheted stretch sensor. This had some slight issues during the process, but also had some clear moments of connection between the two dancers. The issues were more technical aspects – the sensor did not work wirelessly as Processing would not read the bluetooth module. Also at times the computer stopped reading the USB device (which I have had happen with various Ardunio devices at different points, especially when powering off the computer). However, the breath rate of one dancer did in fact change the starting and stopping of the actuators on the other dancer when it was all working.
Finally, we used data sets we had previously collected and created arrays in Processing to trigger the buzzers. First we used a set from one of yesterday’s improvisations with the Polar heart rate. There were some issues with the rate at which the data was sent to the actuators – Processing went through the set very quickly, and usually was done with the first half of the set by the time the OSC connected. This made for some odd latency issues. However, finding a longer set did help with this issue. We also used a data set of respiratory rates from the OMSignal. This was more complicated to achieve as there is no API or web data with this device. Instead, Camille hand recorded all the data from that session in order to be input into the array. What was interesting was that there was a clear difference in how the garment translated the heartbeat versus the breath rate and the dancers were able to translate this into their movement.
Also, the Processing code is going up on github and you can find it here: https://github.com/sicchio/hacking-the_body
Here are more images Camille took of Kate helping dancers with the wearable she made with vibe boards and the wireless X-OSC board that communicates with Kate’s computer:
Kate troubleshooting the sketch she made:
Even though OM Signal said they would try get us some data to work with it was’t in time for our residency so Camille had to hand ‘collect’ the data from some of our sessions by trying to find the peaks and valleys of the heart rate and respiratory rate, so that we could use this data in the array that Kate wrote in Processing, here’s a screen grab of it:
Below are the dancers exploring movement with handmade wearable devices on the neck (with the Flexinol/Muscle wire and pressure sensor) and the arm (stroke sensor and vibration actuator):
All in all the day, was very revealing in terms of giving us some ideas of which direction to go in for future iterations of the project, papers and more funding. More will be revealed in future.
For now, the next step will be to create a custom sensing and actuation garment, maybe work with OM Signal as well for sensing, if they get their API done this year and wish to work with us, and develop some movement vocabulary with dancers that explore a few specific sensors or physiological data, likely with more X-OSC boards with vibration, as the response is very easy to work with in terms of response in movement, and possibly more with muscle wire for other haptic interaction.
Today felt like a lot was accomplished. We finished our DIY garments and wearables and planned activities for the next two days where we will be workshopping at the Siobhan Davies studios with two dancers. We also spent time collecting data from our commercial biosensing products and started our professional documentation process with the lovely Dann Emmons.
The bluetooth module on the breath (stretch) sensor.
This morning, Kate spent time still trying to get the serial data sent via bluetooth from the breath sensor to be read by Processing. It never happened but a USB version does work. When plugged in, Processing takes the serial data from the crotched sensor and then sends this vis OSC to the xOSC and the buzzer garment. A bit more tweaking of the limits of the serial numbers to trigger the buzzers and we will have one DIY garment on one dancer talking to another DIY garment on another dancer.
Here is the wireless buzzer garment and the plugged in stretch sensor.
We spent time this afternoon also collecting data from both the OMsignal shirt and the Polar belt. We created reports of respiratory rate, heart rate and calories burned. This data could possibly be used to create an array for the buzzers, with higher numbers vibrating the left arm and lower numbers vibrating the right (for example). We also plan on using these commercial sensors tomorrow when we work with the dancers and explore research questions in terms of data collection ethics and identity.
One thing that didn’t happen in Day 4 was the connection of Camille’s little muscle wire actuator to a sensor.
So for Day 5, (though she’s wasn’t feeling well), Camille connected a pressure sensor to the neck piece and connected the buzzer to the stroke sensor.
and reconnected the neck band to the 3.7 Lipo battery. The pressure sensor is very simple conductive fabric separated by foam and connect to (+) and (-) reconfigured jumper wires to have a more solid connection for the very sensitive and subtle muscle wire (like a delicate flower) 😉
The rest of the we worked with the OM Signal and Pulsar to get data for Processing with the dancers tomorrow. Camille had a conversation with the makers of OM Signal who said they would try get us the raw data to work with as Kate mentions above, to then find a way to work with it in real time with the dancers.
Pictures of that to come from our documenter Dann Emmons.
Here are the images from our 3rd workshop June 17th, 2013 in Sydney, Australia – on intro to soft circuits this time with intro to Arduino programming, at the 9th ACM Conference on Creativity & Cognition 2013
Here are the images from the TekStar Festival, Hacking the Body – intro to soft circuits crafting for performance, in lovely Byron Bay, Australia, June 15th, 2013 – see also the TekStar website http://www.tekstar.tv
- Rain http://rainycatz.wordpress.com/2012/10/22/baroesque-barometric-skirt/
- Kobakant http://www.kobakant.at/DIY/?p=2542
- Kobakant & Imogen Heap
- Nama – Motion Sensing Interface http://viraseres.com/nama/
- Ying Gao http://www.fashioningtech.com/video/playtime-ying-gao
- Smart Textiles Design Lab Blog
How to Get What You Want
Lilypad and e-sewing basics/Leah Buechley
- electronics for absolute beginners http://www.mztek.org/wp-content/uploads/2010/11/electronics-for-absolute-beginners-r4.pdf
- Basic Electronics http://www.fashioningtech.com/video/video/listTagged?tag=Basic+Electronics
- Soft Circuits http://www.fashioningtech.com/video/video/listTagged?tag=Soft+Circuit+Tutorial
- Beginning Soft Circuits (e-book) project guide http://www.instructables.com/id/Beginning-Soft-Circuits-1/
- Lilypad basics http://www.instructables.com/tag/type-id/?sort=none&q=Lilypad
Soft Circuits Saturdays Wearable projects and resources
Felted Signal Processing http://fsp.fm/
Creating Felt Sensors http://learning.codasign.com/index.php?title=Creating_Felt_Sensors
Materials/ Sensors to buy or make:
Flexible Stretch sensors http://www.imagesco.com/sensors/stretch-sensor.html
Plug and Wear http://www.plugandwear.com/default.asp?mod=cat&cat_id=86
Plusea soft electronics http://www.plusea.at/?p=938
Awesome “kit-of-no parts” electronics you make http://web.media.mit.edu/~plusea/?category_name=actuators
Conductive materials http://3lectromode.com/blog/2013/01/01/conductive-materials/
Arduino (Lilypad & Flora) resources:
http://www.adafruit.com/category/92 – overview of Flora & to purchase
http://learn.adafruit.com/getting-started-with-flora – getting started
http://lilypadarduino.org – getting started
http://www.learning.codasign.com/index.php?title=Beginning_Arduino – more on getting started
http://arduino.cc/en/Tutorial/Foundations – Arduino’s own learning resources
Australian electronics distributors:
Other Key Suppliers (non-Australian)
Adafruit – http://www.adafruit.com
For the Unraveled performance, I have been working with conductive yarn. What I have found is to make sensors with a resistance that is measurable. I don’t focus on the electronics and measurements so much – just what seems to work. A lot of this is based on Kobakant’s How To Get What You Want. So I highly recommend checking them out, especially if you want conductive and resistant measurements.
Basically I crochet a sample of yarn with the conductive yarn and normal yarn just paired together as if I were crocheting with one piece. The conductive yarn is Nm10/3 conductive yarn. The stitch I default to is a double treble (UK) crochet stitch. I actually taught myself crochet through these tutorials.
I have been trialling different types of normal yarn and have found some just are not resistive. My test is a simple 3.7V battery , resistor and LED and when the yarn is stretched. In the image above you can see the types of yarn I have tried.
Silver – All conductive yarn. Great as a conductor but no resistance
Blue – 80% bamboo 20% wool. No resistance.
Pink (used in video) – 100% cotton. Works best.
Yellow – 100% Acrylic. No resistance.
Pink/White mix – 70% triacetate 30% nylon. No resistance.
Hot Pink (pictured below on soft circut sample) – 50% Acrylic 50% cotton. Works well.
I have used the working trials in creating soft circuits with the crocheted stretch sensor, battery and LED with conductive thread. Below the crochet is sewn onto stretch jersey fabric so the stretch comes from the fabric.
And here are some videos of the 100% cotton yarn in action
Recently I have been working on a dance piece called Unraveled which utilises crocheted stretch sensors across the ribs to measure breath. This is then sent via a microcontroller wirelessly to a laptop where it manipulates sounds. During this process there have been many technicalities in working with bluetooth (Bluesmirf in particular) and Flora which have made me change directions in the parts/technology which will be implemented this project.
First idea was to use the Flora microcontroller from Adafruit. This is a newer microcontroller that is for wearables with slight differences from the Lilypad (mostly around the battery). Adafruit plan to release a bluetooth module according to their website. This being the case I thought the Flora might work with other bluetooth modules, such as Bluesmirf. Unfortunately this is not the case.
Firstly, I had to reconfiguire the Bluesmirf anyways. There is more about how to do this here and here. There is also a small bit about this in Tom Igoe’s book Making Things Talk (2nd edition) on page 68. Basically it has to be changed from HID to SPP in order to pair with a laptop and send serial. I did this first in Cool Terminal and later realised I could also do this in the Arduino software. However, even after trialling various CMD codes, it would still not handshake with the Flora.
It is definitely not the Bluesmirf because the set up works with the Arduino Uno. And I can send serial from my stretch sensors to the computer wirelessly. The problem with the Uno is that is is a bulky board for prototyping, not really to be worn as part of a costume.
My next idea is to move away from bluetooth and use xbee and lilypad. This is a working combo and will allow for sewing the lilypad onto the garment that is also housing the stretch sensors.