Embedded Files
Read along to see our progress week-by-week and the progress we made, problems we faced, and plans we expected to finish in the coming weeks!
Week 1:
Progress: We gathered information about our client’s preferences. They shared that they envision the EMG device as a musical instrument or creative tool rather than a game, with a focus on producing actual musical notes instead of game-based interactions. They also mentioned that a visual application could be a possible addition. Additionally, they explained that due to sensory sensitivities, the students require non-stick, wireless electrodes. We also learned that some students can only use their jaw or head, so the design must support EMG straps that can be worn around the head.
Problems: The requirement for non-stick electrodes makes it more challenging to securely attach the sensors to students. Wireless electrodes are also more expensive, which may limit the number of electrodes we can use and reduce the number of musical notes available. While our online meeting was successful, we have not yet been able to schedule an in-person visit to the William E. Carter music therapy class. As a result, we do not yet have a concrete understanding of how to design for the students’ specific and varied disabilities.
Plans: We plan to place our first order for wireless electrodes this week. We also plan to experiment with the MyoWare conductive sleeve, which is compatible with electrodes and may provide a comfortable attachment option for students. Additionally, we will explore fabric straps as another method for attaching electrodes. We also plan to determine how to configure the electrodes using a Raspberry Pi.
Week 2:
Progress: We ordered the MyoWare 2.0 muscle sensor and selected the ESP32 as the main microcontroller due to its built-in analog-to-digital converters, which simplify EMG signal acquisition without requiring external ADC hardware. With these components selected, we began initial system integration planning. An Arduino-based test program was written to read raw EMG signals from the sensor and process them on the ESP32. While this code has not yet been physically tested, it is prepared for upcoming validation once the components arrive.
Problems: We have not yet been able to schedule an in-person meeting with our clients, which has slowed down progress on finalizing the design. In addition, we do not yet have the wireless attachment for the MyoWare muscle sensors. While we have the basic sensors, we are still working on determining how they will be fully integrated with the software.
Plan: As the SparkFun components arrive, we will test each part and determine which materials will be used in the final design, since not all shields included in the kit may be necessary. We have also scheduled a meeting with our clients to better understand the environment where the device will be used and to ensure the design meets safety and usability requirements.
Week 3:
Progress: We ordered conductive fabric to begin prototyping how the device will be physically attached to the user. We visited the school and observed how the kids participate during music classes, which gave us a much better understanding of their movements, engagement, and practical constraints for wearing the device. We also tested the existing ESP32 code with the MyoWare muscle sensor, and it successfully detected EMG signals from muscle activity. The program now displays the EMG signal in real time as a graph.
Problems: We are still waiting for the wireless shield needed to test the wireless connection, which limits our ability to evaluate the full system communication. In addition, the EMG signal is currently unfiltered, so noise is still present and must be addressed before reliable note detection can be implemented. The mobile app has not yet been started, so integration between the hardware and software interface remains incomplete.
Plans: Now that we have a clearer understanding of the kids, their environment, and how they interact during music classes, we will focus on:
prototyping the wearable design using the conductive fabric
determining the best way to integrate the sensors comfortably and securely
On the software side, we will:
implement signal filtering to clean the EMG data and improve reliability
begin development of the app
start defining and tuning the EMG thresholds required to trigger each musical note
Week 4:
Progress: We successfully connected the output of the EMG to playing a note while ignoring noise. This note is currently being played by a buzzer. The device was shifted from being usable by hands and feet to being controlled by jaw and neck movements. This change was implemented after a visit to the client and a better understanding of their conditions and needs. The Myoware kit also arrived which contains the shields required to organize the device and wires.
Problems: We are unsure how to make the device attach to the user in a comfortable but secure way, as the device will be directly on their face/neck. We are also still unclear on the direction of our final note output system, and whether a digital interface or real keyboard may be more beneficial and doable.
Plans: Now that we have refined our product use to neck movements, we will:
- Research optimal attachment options
- Finalize our note implementation device
- Start developing casings for the Myoware for cleaner interface
Week 5:
Progress: Because of many troubleshooting issues, we have not made much progress from last week. As we continue the design process, we have found that the power shield is the most effective way to connect the Myo ware sensor to the Arduino; however, this has made our connection to the sound buzzer to be very difficult. Tfherefore, we will need to temporarily remove the power shield in order to create a sound output for our MVP.
Problems: We are still thinking about what the final result will look like and how the EMG signal will output music. We do not know what format the music will be in or how it will be played. There is no space left to connect a buzzer since we are using the Myoware power shield, and when connected to the Arduino, that leaves no pins available. We also do not know how to place or secure the electrodes without using sticky pads, and are considering using a band to hold them in place.
Plans: As we continue to troubleshoot the wired connections between the device, extenders, and electrodes, we will:
Change from a buzzer to a computer-based audio
Finalize which shields we will be using in the final product
Finish implementation of the device in the sleeve
Week 6:
Progress: We were able to reconfigure our design to function with the basic buzzer. This allowed us to present our MVP to our clients, who were excited to see the direction the project was heading. While the shields that we originally planned to use proved to be ineffective, we ordered a new electrode kit that may solve our problems regarding signal acquisition. We also finalized the electrode site as the neck, right under the ear, rather than the jaw. This allows for consistent thresholds and a wide range of potential uses, such as various thresholds or time-based output control systems.
Problems: We are still unclear about what the final result will look like and how the EMG signal will output music. We do not know what format the music will be in or how it will be played, as there are very limited options with one or two muscles. Also, if the electrodes ordered do not work, we will struggle to secure a functional signal in a sustainable, reliable way.
Plans: As we continue to troubleshoot the wired connections between the device, extenders, and electrodes, we will:
Finalize the implementation of the device in the sleeve
Identify the objective of our program; i.e. game, music-making tool, etc.
Begin a digital interface that plays a note from a computer based on thresholds
Week 7:
Progress: We have created a basic program that can play 6 different notes.
After the new electrode kit failed, we are now focusing on making the project work fully with the sticky electrodes before tackling the problem of using dry electrodes and making the project wireless.
Problems: After the new electrode kit failed we are unsure as to how we are going to resolve the client needs of dry electrodes. We are also struggling with receiving a signal reliably and for extended time before the system stops functioning. Previous casing solutions have also failed as a result of the dry electrode failures.
Plans: As we continue to troubleshoot the wired connections between the device, extenders, and electrodes, we will:
Continue to develop the audio display program
Figure out a way to reliably receive electrical signals
Start brainstorming for new potential dry electrode solutions such as bar electrodes
Week 8:
Progress: We have changed directions on wireless electrodes.
We are working on using reusable silicone pads that are wired directly to the hardware and consistently provide accurate data.
Problems: Our current main problem is the cable length and finding a 2.0 mm male DIN to snap lead cable that is lengthy enough. Another problem is that the reusable silicone pads have a limited reusability and it would be a reoccuring cost for the school.
Plans: As we continue to develop the code and app, we are thinking of ways to
provide multiple, easily changable audio options
define electrical thresholds within sets of notes
Week 9:
Process: This week, we focused on putting the system together as a more complete and polished product. We soldered a connector that allows the client to use either the bar electrodes or the reusable sticker pad electrodes, which gives more flexibility during testing and use. We also 3D printed a black box and have been assembling all the hardware so the system looks cleaner, more organized, and more professional. At the same time, we have been working on the build-it-yourself guide, the user manual, and the UI.
Problem: The main issue right now is that the bar electrodes are still barely working, so they are not yet giving us the level of performance we want. Because of that, we are still relying on different electrode options while figuring out what works best. Another challenge is making sure everything comes together in a way that is not only functional, but also easy for the user to understand and use.
Plan: As we wrap up the final details of the project, we will:
Finalize the wiring
Ensure the casing is functional for all aspects intended
Week 10:
Process: We spent this week finalizing the last details of our presentation. We finalized the wiring and are finishing up the casing.
Problem: We were unable to make the bar electrode functional, so we are sticking to the resuable sticky pads rather than giving two options.
Plan: As the project comes to the end, we will:
Present the project to our clients
Page updated
Report abuse