Final Iteration - Brainstorming and Lock In - Hasan Berk
Big Heart SETU
The Final Iteration 🧠🔒
Alexa, cue 'The Final Countdown'
Brainstorming ðŸ§
Now that all the technical work was done, in relation to calculating HRV, the MQTT and IFTT work, it was time to brainstorm ways to bring it all in together to create a powerful visualization bio-feedback of HRV to the end user.
One suggestion that caught everybody's attention was the 'ship in a bottle' toys that rock around without ever sinking. Comparing your HRV to a stormy sea when low, and a nice calm sea when high as your heartbeat slows down sounded like an amazing idea.
 |
A Calm Sea representing high HRV |
 |
A Stormy Sea representing low HRV |
This led to us researching through YouTube and finding a whole community on these, specifically focusing on something called Wave Machines. These are essentially the same concept, and is a rectangular box with 'waves' rocking back and forth. When we stumbled across this particular wave machine, our eyes lit up.
This video was so mesmerizing, that we finalized the decision that we need to build something similar, and to integrate it into our HRV setup. It is a good symbolization of both HRV, but also has a calming effect, which positively impacts your HRV.
Myself, Jack, and Adam started researching ways we can build something similar. Unfortunately the machine shown in the video is the result of 10+ years of research as stated on the website from the creator. "This handmade representation of modern kinetic art is the result of almost 10 years of research, testing, and artistic exploration by Hughes Wave Motion Machines LLC, a Philadelphia-area small business" Reference. This made the machine in question slightly out of scope, however the biggest lesson from four years of computer science is how to problem solve. We decided that we can build something smaller, and therefore scoped out a smaller version. We researched local options to possibly modify and found this version.
However we later decided to create our own version of the wave machine. This decision was driven by multiple purposes: to gain hands-on learning experience , to thoroughly understand the internal mechanisms of such a device, and to not limit our possibilities in aspect to the design.
To ensure everyone had adequate work, the other team worked on a second form of bio-feedback. Their idea was a teddy bear with a breathing effect. This teddy bear idea not only provides comfort and companionship to the child, but is also an emotional approach to bridge a connection between a child's heartbeat HRV with their parent. By providing a physical manifestation of the child's physiological state, it too is a powerful way to visualize HRV. The teddy bear that was decided to be used for their project can be found here.
To complete the brainstorming, we decided to have a two-day lock-in to be able to narrow down our entire focus to development and wrapping up the 'Big Heart Project'. The idea of a lock in was appealing to us all in our final semester, as the workload was rampant and having it after all assignments was a great way to focus our entire energy into this module. We scheduled the lock-in for the 7th - 8th of May from 10pm - 4pm. This 12 hour window was allocated for the entire building and testing of our bio-feedback devices.
Now that the teams were decided, we had to split tasks up. For this I suggested a collaboration board called 'Miro' which I had used during my internship in semester six. Miro was a great way to create a list of tasks and define them together all in one place - kind of like an online whiteboard.
 |
Now that the work was scoped and defined, it was time to wait for the lock in 🕒. |
The Lock In 🔒
From our brainstorming session, we had a list of tasks we could choose for this project. Keeping the two-day window in mind, we decided on the following:
- Build a wave machine using servo motors.
- Create a wearable to use with a different Microbit and communicate the HRV values from this to the wave machine's Microbit.
- Polish up HRV Code
- IFTTT implementation
Day One
On the Tuesday morning myself and Adam arrived early to begin working on the wearable which is the starting point of our design. After rooting through some designs on 'Thingiverse' , we began printing some designs of wearables. Since Adam is more experienced with 3D Printing, he showed me how it works, and the process of printing it. Once the wearables were printed, we filed them to smooth them for a finished look. Here is a video of both wearables. The bracelet was not as accurate so this was scrapped in favour of the hospital style enclosure.
We had two initial prototypes of smaller scale 'wave machines' made using water bottles, baby oil and water with food colouring. These are shown below.
Myself, Jack and Adam decided to take another trip to shops to see if we could find anything more rectangular and narrower, to mimic bigger waves, yet still be light enough for our servo motor to be able to move comfortably. Eventually, we found a bottle of shower gel which was close enough to what we needed.
We had a couple of problems to solve, mainly with the stress the servo was under. The original plan was to implement a balancing system with the servo pushing one side of the bottle up to rock the container. This way, using one servo we could still get the container to rock. However, due to limitations with the strength of our servo, we had to think of something else. I noticed some old fishing line in the lab, and suggested perhaps suspending the bottle, and use pulleys to offset the weight of the container. Pulleys are a fascinating way to offload weight and distribute it better.
We took some old cardboard from the lab and built an initial prototype. Myself and Jack began building the cardboard stand, using parcel tape to hold it together, and nuts and bolts as holding stands for the container. In the meantime, Adam was working on Microbit code to manually control the servo for testing purposes. Alas, we completed the first prototype of the wave machine, and our idea was working!
 |
| Manual Testing Code |
Although it wasn't pretty, our proof of concept was working. The makeshift pulleys using the bolts resulted in much smoother movement of the container as some of the weight was shifted off of the servo.
Day one's result was a proof of concept working, and the next day's goal was to finalize the design and bring an end product of the bio-feedback wave machine.
Day Two
With our success in day one, we were extremely enthusiastic walking into the lab on Wednesday morning, ready to improve our prototype. To our disappointment, we were met with a new unforeseen issue. The fishing line had stretched overnight under the tension of the pulleys and the weight of the container, rendering our prototype now useless as we had to re-attach fresh wire.
On the way to the lab that morning I noticed a possible solution for the final design of our wave machine - a box I have walked past every time coming into the class. It wasn't the box that was special however, but the old ring-binder folders in them. These folders have been sitting there as trash for a while, and the shape is the exact same as the cardboard prototype we had made the day prior. After getting permission from Jason, I took one to begin building a better prototype. Using these old ring-binders also add an element of sustainability to the project, recycling them for a new purpose. I threaded in bolts in the corners to keep it closed, and drilled holes to fit the existing bolts into.
While I was making this, Adam 3D printed actual pulleys, to give our final product a cleaner look. We had to use tape to act as a friction reducer within the pulleys, to not damage the fishing line. When we re-implemented the original prototype's design, we decided to add another pulley for further efficiency. We used another folder as a base, which was a box shape, to prevent the folder from toppling over and to evenly distribute the weight of the container. In the meantime, Jack created an IFTTT implementation to push HRV values to a Google Sheets document.
During lunch, we went to Tesco to get some string to fix the problem with the fishing wire stretching. Implementing this into the design, and adding some aesthetic 3D printed versions of the logo I made for our project, we had a nice looking final prototype.
Now that the bio-feedback device was complete, we all started working on getting our HRV values to the wave machine Microbit. Both teams only had access to the one shared cloud board, so we had to constantly modify it to test our changes. It took all of us with constant debugging and testing to reach the end code, but it was eventually a success.
We altered some of the cloud code to fit both teams goals, which was a team effort on both sides. The code can be found below:
The Wave Machine code uses the HRV values received to map them to Servo angles, running in a forever loop with one second intervals:
Finally, the code from our Microbit that takes in the HRV from the pulse sensor and sends these values to the MQTT using the Cloudboard is found below:
To tie it all together, we successfully had a working Wave Machine, powered by your HRV. The full video demo of me using the wave machine is show below:
Conclusion
Over the course of the semester, we gained invaluable knowledge on HRV, IoT concepts, MQTT, system design and architecture, Microbits, Code Blocks and Team-Work. I am extremely grateful that this project was a success, and we managed to work together as a team to build the final product, with each iteration contributing more and more to the end goal.
I would like to thank Jason for the welcoming classroom and his extremely helpful method of learning. It was nice to have a module where we could unleash our creativity and drive a project, working as a team. I would also like to thank all my classmates for their contributions through the development of this project.
Comments
Post a Comment