This week the project has really moved forward – and fast! The most important milestones / achievements this week was:
- I received the NeuroSky Mindwave Mobile Headset – and I was able to receive JSON data through a JAVA socket!
- Also I got introduced to the DUL Radio, and even though it was a little harder to connect (mostly because I was very slow to realize how to set it up) I was able to create a JAVA class that parsed the incoming data.
- With the help from Mads Møller, I now have a loosely coupled software system, which has already proven very effective and easy to expand. I already have three different types of connections: EEG = socket, DUL Radio = special made JAVA connection, Exiting Arduino Bluetooth prototype = “normal” serial connection. On top of that they all use different protocols 🙂
- I Improved the GUI interface. Right now I use a tap-layout, which was relatively easy with the ControlP5 library (never really used CP5 before, but very powerful for standard features)!
- The GUI interface is able to show the activity of all sensors at the same time. It is still possible to switch to the “Raw sensor view” and see the raw sensor data for all connected sensors.
- Morten Boye Mortensen – one of the best (iOS)developers and designer I know, programmed an iOS app for me that can retrieve the RR values from my Wahoo Fitness Blue Heart Rate Strap, which I can use to calculate the Heart Rate Viability.
- 5 pages of related work in short resumes of 14-18 papers.
On the downside, I am still very insecure on my research question and the relevant related work / areas. My meeting with Kaj didn’t bring me much closer to a better research question, but it was a big help that he thought I did very well and that he was sure there were a lot of interesting research questions / paths to explore. Also, it is easier to make a plan and a research question after my meeting on Thursday, because then I’ll know if I’ll be able to test my system with children with ADHD (fingers crossed)!
My goals for next week are:
- Create two functional prototypes with embedded sensors that I can demo on Thursday.
- Improve / calculate / calibrate the sensor data from the DUL Radios in JAVA.
- Make a really really good and convincing slideshow for Thursday.
- Create a prototype (Arduino, Bluetooth, Accelerometer+Gyro) for Mads and his PhD project.
- Talk with people that knows about Sensor Fusion and machine learning, so that I know how I should sample, calculate, record etc. data from my experiments.
- Write 2-3 pages of related work.
- Find a way to visualize the EEG data in the GUI (a new tab?)
My week in pictures (random order):