Posted on

Sydney’s Status Report for April 8

This week I worked on adding quaternion animation to the website. This worked successfully and now we can see the quaternions over different points in time visualized. This is still on a fake data set however.

 

Next, after the new arduino board arrived on Friday I got to work on the bluetooth with the on board quaternion processing and was able to begin transmitting one characteristic of data to the web service. I would call this a very succesful week for me, and am on track. Next week I am going to work to have the data stream to the quaternion visualizer be our raw data, as well as adding a trigger on the website to tell the device to start recording and sending the quaternion data.

 

To ensure my portion of the project meets our specification I intend to create the visualizations of the quaternions and visually confirm that the sensor readings match the relative position of the back. I might user test this on multiple different people to ensure it matches across multiple user profiles.

Posted on

Zhichun’s Status Report for Apr.8

This week I focused on building the circuit on the feet, and I think my progress is on track. This circuit is supposed to be able to detect pressure. Our team chose to accomplish this by building a pressure sensor, and I built a voltage divider with the sensor and a resistor. With the different amounts of pressure, the resistance of the pressure sensor will be changed, and by measuring the voltage in between the pressure sensor and the resistor, I will be able to get the amount of force applied to the pressure sensor.

My contribution to this project is the physical part of the vest and the circuit of the lower pressure system, I will analysts and evaluate them by seeing if they are functional.

For the next week, I plan to implement the code for the pressure sensor with Arduino, and I found an article online explaining a similar subject to this.

https://srituhobby.com/how-the-voltage-sensor-module-works-with-arduino/

Posted on

Team Status Report for April 1

As stated in the previous status reports, our biggest risks are our algorithm not working correctly and our Bluetooth not communicating correct data. To mitigate these risks, we’ve been doing lots of testing to ensure the data read from our sensors is correct and researching quaternion libraries to help with our algorithms. There have been no changes to our design currently, however, some members seemed to encounter enormous difficulties when completing their tasks, so we are currently behind schedule. In order to account for this, we’ve decided to shorten the number of days allocated for testing, without reducing the total time spent and also re-assign some tasks. By doing this, we hope to still finish on time before the final presentations.

Our updated schedule
Posted on

Jasmine Yew’s Report for April 1

This week, I primarily researched quaternion libraries and how to use the generated quaternions to generate feedback. I’ve decided to use an external Javascript library called three.js and to generate feedback, I plan to use Three.js to generate the vectors to display on the feedback page. Since it seemed like no progress had been made on the tracking page, I also created an interim tracking page with a basic HTML layout where users choose the exercise they want to perform, then get directed to a new page where they can begin tracking their set or stop tracking their set. I am slightly behind schedule right now because my parts of the project rely on parts that my teammates are still working on and have yet to complete. In order to keep on track, I’ve been using fake data and plan to keep using fake data to test the quaternion generation. By next week, I hope to have the quaternions generated, hopefully using the data received from the sensors.

Posted on

Sydney’s Status Report for April 1

This week I mainly worked on optimizing our libraries so we could both use the onboard DMP processing to get quaternions as well as the arduino BLE library. This was a real problem as our program as previously written was using 120% of the program memory space. Unfortunately due to problems around that we have yet to fully test the bluetooth integration(but this will hopefully happen tomorrow before the interim demo). But after spending a couple hours better understanding the libraries and removing functionality unused by our application I was able to get our program memory usage down to 99%. This progress is visible on our github. I also worked with Rachel a bit to decide on how to sew the devices together and make sure it matches our vision. We are just barely behind as we were hoping to have the data read to the web server by now, but I think after tomorrow that will be done, and then next week I will dive into working on the foot sensors and assisting Jasmine as needed on getting the detection algorithms working.

Posted on

Zhichun Zhao’s Weekly Status Reports for Apr.1

This week I focused on the physical level of the project. I designed and built a wearable system with expandable straps and buckles that fit the accelerometers, the Arduino, and the power banks. I used the sewing machine along with the straps that I ordered at the beginning of the semester, but since the machine stopped working after the first 20 minutes of use, I had to sew the sensors on to the bands manually. This made the process more time-consuming and took longer than I expected. Fortunately, I completed the wearable part by the end of the week.

For the next week, I will start focusing on the tracking page as well as the demo with my teammates.

 

Posted on

Zhichun Zhao’s Status Report for March.25

This week, I focused on two parts. The first part is the tracking page. Since I have not been coding for websites for a year now, I did some research on tutorials that can guide me to build a timer, and I found https://www.youtube.com/watch?v=49f1cjZWRJA to be a good fit for what the tracking page needs. I also follow the django guide to set up the virtual environment and downloaded the newest version of django on my laptop, as now I am prepared to implement the actual timer. I hope to get a working timer next week for the tracking page.

The second part is the physical hardware. I am planning to sew the sensors onto the wearable fabrics during the lecture time on Monday. In order to do this, I went to Techspark on Sunday and learned how to use the sewing machine in the workshop. I also learned how to sew with needle and thread for more delicate or smaller parts of the sensor. I believe my skill is prepared to satisfy my teammates’ requirements and I will be able to complete the sewing process on Monday.

Posted on

Jasmine Yew’s Status Report for March 25

This week I primarily worked on generating potential feedback and tips for users’ squats and deadlifts. Since we do not use databases or machine learning, I’ve decided to hardcode the feedback. I also created the profile page for our web app, in case users want to update their weight and information.  In addition, I began looking into potential ways to receive and send data to the Arduino via BLE and the web app. Currently, I am on schedule, however, I need to test the Bluetooth functions. Hopefully, by next week, I can have the Bluetooth functions fully working such that we can create the Quartenion vectors after.

Pseudocode of how we will be generating the feedback.
Posted on

Team Status Report for March 25

At this point in time, the biggest risk is that our device won’t fit nicely onto our fabrics, and that our algorithm does not work.  This is because at this point we have managed to get the IMUs working individually, as well as developing most of the individual website pages. We will be mostly on track for the interim demo once we get the sensors attached to the fabrics on Monday, as well as after we refine the bluetooth to web app connection. I don’t believe any changes were made to our system this week, we have mostly just all been working on our individual parts to get them ready for the demo in a week.  I think our goal for the interim demo will be to showcase the data collection from the back unit to the web application.

Posted on

Sydney’s Status Report for March 25

This week I primarily worked on getting the IMUs working with the mux. I ran into a lot of problems with the IMU’s not being detectable once connected to the MUX, but after troubleshooting for several hours, realized that for whatever reason with the MUX, I was only able to read the sensor data if I decreased the clock speed for I2C to 100kHz. So I am officially ‘done’ with getting the back device working / reading data. My main goal for the upcoming week is going to be getting the bluetooth from the arduino to our webapp working so that the users’ data can actually be displayed / processed. Luckily, I am actually mostly caught up now. I still haven’t made any progress on the foot device, but after working with Jasmine to get bluetooth working I will get started on that. We were also able to get a bluetooth example working, so I am hopeful that converting the example to our application specific use case won’t take significant time.  Unfortunately because of the issues with the MUX, our placing sensors onto fabrics was delayed(as we wanted to make sure everything was working before committing to them), but that should happen without much problem next week) So in effect next week I am hoping to get the sensor data communicated via BLE to our webapp, and I am going to work with Rachel to get the sensors attached to fabrics so that it can be ready for the demo. I will likely also experiment with making sure the batteries we got provide sufficient power to our device.