Along with a collaborator from MIT,(1) I helped create rSkin, a multi-touch stretchable textile sensor for robotics.(2) Here are some really interesting things

Along with a collaborator from MIT,(1) I helped create rSkin, a multi-touch stretchable textile sensor for robotics.(2) Here are some really interesting things I learned about e-textiles during my project - The hard/soft interface is really hard. At some point you need to interface with electronics and in many cases this will be a failure point. Also, normal electrical contact methods like soldering just don't work. - Stretch is really hard. Most of the fabrics shown don't look like they have much stretch, but I wanted something that could wrap around robot joints. Turns out that calibrating a sensor for this task is very hard to model. Even without stretch calibrating and maintaining calibration can be hard. - Metal oxidizes I'm hoping Google has taken care of this but many of the thread types I had at the time were silver based and would degrade (become much more resistive) over just a few months. - Getting high resolution sensors is tricky. Getting a ton of conductive threads wired up in a matrix means you're trying to do lots and lots of hard/soft interfaces. Automating bringing all those threads from one grid direction into a tight layout to match with a board or other metal conductor will be a very interesting manufacturing challenge. I'm surprised and excited to see Google take this on as I still feel there is huge potential for e-textiles in softer and more sensitive robotics. (1) Hannah Perner-Wilson (who did all the hard work) (2) Check out the instructable (http://www.instructables.com/id/rSkin-Open-Source-Robot-Skin/ http://www.instructables.com/id/rSkin-Open-Source-Robot-Skin...)