Rahul Nair







Level of Interest function
Counter activity
Dispatcher's rescue
Everywhere display
Less is Moore



Counter Activity

Tracking user activity using RFID


Counter Activity was a class project for Mobile & Ubiquitous Computing (CS 7470). We began the project with the aim of tracking user activity at a kitchen counter. This was inspired by research on Deja Vu displays that is currently being conducted at the Aware home. While the Deja Vu displays currently use a wizard of Oz system to detemine user activity on the counter we believe that it will eventually be possible to automatically gather activity data

Design and development process
We went through an extensive brainstorming to decide exactly what type of sensing technology would be best suited to our goal. Some of the technologies we considered were vision, touch sensitivity, ultrasonic trackers and accelerometers. We eventually settled on using RFID as our primary tracking technology since it was a cheap and robust system as well as the fact that many industries are planning to add RFID tags on consumer products which will allow any RFID based tracking system to quickly acheive critical mass.

Most current RFID systems involve a fixed RFID reader and require that the tagged objects be moved across the readers field to be read. We wished to reverse that trend by attaching the RFID reader to the users hand while tagging the various objects that they would come in contact with. By constantly polling the RFID reader we would be able to determine which objects the user was interacting with. We also planned to embedd RFID tags in strategic area on the kitchen counter which would enable us to simultaneously track the position of the users hand.

After receiving an RFID reader we quickly realised that not only was the reader extremely large, but it also had a very low range (1-3 inches). This meant that we had to investigate alternative technologies to track the position of the users hand. Our first choice was the HP badge 4 system since it not only had an onboard accelerometer but also a strong ARM processor which could be used to control the RFID reader as well as wireless capabilities which would allow the users to work without a tether. Our next choice was to use a Nintendo power glove which used an ultrasonic tracker to determine hand position and orientation with 6 degrees of freedom. It also had an elastic sensor which could sense if the users hand was open or closed. We purchased a powerglove but were unable to hack the Nintendo interface to a computer.

Eventually we a software system that could accurately sense and keep track of the various objects that were in the RFID readers' range. We also used Wizard of Oz inputs to allow the system determine the position and orientation of the users hand.

Project reports
The original swiki that we used for the Counter Activity project can be found here.