Experimenting with EMG

In the upcoming issue of ACM interactions magazine I will talk in my column on interaction technology about the electrical interfaces to humans. This blogpost is written to give you a preview of the article and describes a little project I did over the weekend to explore Electromyography (EMG).

It is the simplest circuit I could find and it results in a very simple and easy to make EMG. It takes less than half an hour to build and plugs into the microphone input of a computer. It uses a single operation amplifier (INA128p) and is based on the description in [1]. Builing the filter into the gain branch is quite nice ;-)

For my Experiment I used the following components:

• a INA128p (Precision, Low Power Instrumentation Amplifiers)
• 2 Capacitors (100uF, Elco) 
• a Resistor of 120 Ohm 
• 3 copper pieces (e.g. 5 cent coins) as electrodes for the body 
• a 3.5mm audio connector (from broken headphones) 
• some cables and prototyping board 

The eagle CAD file is available for download. 
Connecting it to other parts of the body made a very simple ECG – more suitable to get the heard rate that the shape of the signal.



More sophisticated circuits are available, e.g. openBCI [2] or an EMG shield for Arduino [3]. If you look more for commercial device you will find in [4] a good overview and also an introduction to emotion sensing.
 

[1] Bhaskar, A, Tharion, E., and Devasahayam, S.R. Computer-based inexpensive surface electromyography recording for a student laboratory. Advances in Physiology Education 31, 2 (2007), 242–243.
[2] http://www.openbci.com/
[3] https://www.olimex.com/Products/Duino/Shields/SHIELD-EKG-EMG/ 
[4] Kanjo, Eiman, and Alan Chamberlain. "Emotions in context: examining pervasive affective sensing systems, applications, and analyses." Personal and Ubiquitous Computing: 1-16

Floor for activity recognition

Patrick Baudisch showed in Paris at the Microsoft Software summit interesting photos of their DIY activity in creating a glass floor for tracking and activity recognition. With a fairly large glass pane in the floor they have created an interesting environment… I am sure there will be interesting things coming out of this installation.

Some years back in 2002 (and looking at the photos and the amount of hair I still had this seems long ago) in Lancaster we also looked in what to do with floors (and we were into DIY as well). We also considered arrangements with a floor tables and furniture on top. As you can see from Kristof, Hans and me on the photo it was a fun project.

The positive point in using load sensing is that you can track unobtrusive and potentially large scales with little instrumentation. We even considered the possibility to put a house on 4 load cells and do activity recognition based on this. We never got around to building the house ;-) The problem with load sensing is that you can only track one moving object/subject at the time.

Looking at the signature of the load measured and doing some signal processing we could detect events – unobtrusive and cheap – but only for single events.

Interested in more details? Have a look at the publications on load sensing [1], on the interaction [2], and at a patent [3] describing the basic technology.

[1] Schmidt, A., Strohbach, M., Laerhoven, K. v., Friday, A., and Gellersen, H. 2002. Context Acquisition Based on Load Sensing. In Proceedings of the 4th international Conference on Ubiquitous Computing (Göteborg, Sweden, September 29 - October 01, 2002). Springer-LNCS, London, 333-350.

[2] Schmidt, A.; Strohbach, M.; van Laerhoven, K. & Hans-W., G. 2003. Ubiquitous Interaction -  Using Surfaces in Everyday Environments as Pointing Devices, Universal Access Theoretical Perspectives, Practice, and Experience (UI4ALL 2003), Springer LNCS, 263-279.

[3] Schmidt, A., Strohbach, M., Van Laerhoven, K., Friday, A., Gellersen, H-W., Kubach, U.; Context acquisition based on load sensing. US Patent 7434459. US Patent Issued on October 14, SAP AG (DE), 2008

Modular device - for prototyping only?

Over the last years there have been many ideas how to make devices more modular. Components that allow the end-user to create their own device - with exactly the functionality they want have been the central idea. So far they are only used in prototyping and have not really had success in the market place. The main reason seems that you get a device that has everything included and does everything - smaller and cheaper... But perhaps as electronics gets smaller and core functions get more mature it may happen.

Yanko Design has proposed a set of concepts along this line - and some of them are appealing :-)
http://www.yankodesign.com/2007/12/12/chocolate-portable-hdd/
http://www.yankodesign.com/2007/11/26/blocky-mp3-player-oh-and-modular-too/
http://www.yankodesign.com/2007/08/31/it-was-a-rock-lobster/

Buglabs (http://www.buglabs.net) sells a functional system that allows you to build your own mobile device.

Being creative and designing your own system has been of interest in the computing and HCI community for many years. At last years CHI there was an paper by Buechley et al. [1] that looked how the LilyPad Arduino can make creating "computers" an intersting experience - and especially for girls.

[1] Buechley, L., Eisenberg, M., Catchen, J., and Crockett, A. 2008. The LilyPad Arduino: using computational textiles to investigate engagement, aesthetics, and diversity in computer science education. In Proceeding of the Twenty-Sixth Annual SIGCHI Conference on Human Factors in Computing Systems (Florence, Italy, April 05 - 10, 2008). CHI '08. ACM, New York, NY, 423-432. DOI= http://doi.acm.org/10.1145/1357054.1357123