Challenges
We've changed our design for this project a couple time during the development. And we would like talk about it a little bit.
In the original design, the communication between micro-controller and Edison board should be implemented in Bluetooth 4.0 Low Energy. This was a nice design since we knew that already we don't want wires going from the inside of your fridge to the Edison board on the outside. The original plan was using the embedded BLE module in Edison and connecting a separate BLE module to the micro-controller (the communication between micro-controller and BLE module would be SPI interface). However, we found that the BLE module we bought was way too small. It's might be nice for a hardware design, but our soldering skill can not handle the small pins on the module. Actually, after trying, we destroyed our BLE module.
And also, in the original design, our egg tray was expandable, which means user can decides the size of the egg tray as they want. However, after user study (observation, mostly), we found that this is totally unnecessary as most people are pretty much satisfied with the egg tray size right now ( 1~1.5 dozen style). All we need to do is to build it larger for users. However, our prototype would still be holding 6 eggs demonstrating the feasibility of our design.
In the original design, the communication between micro-controller and Edison board should be implemented in Bluetooth 4.0 Low Energy. This was a nice design since we knew that already we don't want wires going from the inside of your fridge to the Edison board on the outside. The original plan was using the embedded BLE module in Edison and connecting a separate BLE module to the micro-controller (the communication between micro-controller and BLE module would be SPI interface). However, we found that the BLE module we bought was way too small. It's might be nice for a hardware design, but our soldering skill can not handle the small pins on the module. Actually, after trying, we destroyed our BLE module.
And also, in the original design, our egg tray was expandable, which means user can decides the size of the egg tray as they want. However, after user study (observation, mostly), we found that this is totally unnecessary as most people are pretty much satisfied with the egg tray size right now ( 1~1.5 dozen style). All we need to do is to build it larger for users. However, our prototype would still be holding 6 eggs demonstrating the feasibility of our design.
What Can be improved
After discuss, we think the Intel Edison board we used for our prototype was an overkill. Sure that Edison is a powerful platform and it is nice to have. But the ~$110 cost for Edison board is killing our product to be used by people. It's the same dilemma as the previously mentioned pre-existing product. After exploration and research, we found that we can use some SoC controllers with much lower cost to achieve the same objective. For example, the PSoC ARM based microcontroller by Cypress (~$10 cost), or even an Arduino Uno could do the trick.
However, the overall structure won't be changed that much, excepting we should really replace the wire based I2C communication with BLE 4.0 communication. The uploader unit would still be constantly powered with an outlet based power supply for the reason WiFi really needs power.
To really commercialized this design, we need to take a lot of efforts on user interface, making it easy to use and looks nice. Our current APP is just same as our hardware implementation: a prototype.
Moreover, we could change the interface from APP to Web. As this gives the ability to be used cross-platform, and also for the reason people are using different smartphones other than iOS... like Android, Windows Phone, etc. Developing App for each platform will certainly increase our overall cost. In that sense, switch from App to Web is reasonable for us. Plus, we can also use some data visualization for the histogram of the egg consumption taking advantages of the powerful APIs online for Web.
After all, it's now the age of web.
However, the overall structure won't be changed that much, excepting we should really replace the wire based I2C communication with BLE 4.0 communication. The uploader unit would still be constantly powered with an outlet based power supply for the reason WiFi really needs power.
To really commercialized this design, we need to take a lot of efforts on user interface, making it easy to use and looks nice. Our current APP is just same as our hardware implementation: a prototype.
Moreover, we could change the interface from APP to Web. As this gives the ability to be used cross-platform, and also for the reason people are using different smartphones other than iOS... like Android, Windows Phone, etc. Developing App for each platform will certainly increase our overall cost. In that sense, switch from App to Web is reasonable for us. Plus, we can also use some data visualization for the histogram of the egg consumption taking advantages of the powerful APIs online for Web.
After all, it's now the age of web.
Results
As tested in real life situation, our design works.
As the nature of our design's targeted usage environment, we don't really have huge data to show
Please refer to demostration
As the nature of our design's targeted usage environment, we don't really have huge data to show
Please refer to demostration
WorkLoad
Overall Design - all of us
Hardware Implementation - Zou, Yanbo
Python Programming & Cloud Datapath Implementation - Wang, Yingqi
iOS App Development - Jin, Lingqiu
Hardware Implementation - Zou, Yanbo
Python Programming & Cloud Datapath Implementation - Wang, Yingqi
iOS App Development - Jin, Lingqiu
Conclusion
This project was pretty much successful, despite a few challenges, we eventually made it work.
The skills we learnt from the lecture and lab section of this course gave us a lot of aid. And we also found that the best design is not only handled by skills, but by the idea with the consideration of feasibility. And that what we really learnt from this project.
The skills we learnt from the lecture and lab section of this course gave us a lot of aid. And we also found that the best design is not only handled by skills, but by the idea with the consideration of feasibility. And that what we really learnt from this project.