This project was inspired by the Professional Engineering course taken by all first year engineering students at McMaster University (1P03). The final project for the course was to design a solution to a problem of your choice that was given by St. Peter's Residence at Chedoke, a long term residence care home located in Hamilton, Ontario. One of the projects proposed by St. Peter's was to create a falling alarm to notify the nurses in the event of one of the residents having fallen.
What it does
It notifies nurses if a resident falls or stumbles via a push notification to the nurse's phones directly, or ideally a nurse's station within the residence. It does this using an accelerometer in a shoe/slipper to detect the orientation and motion of the resident's feet, allowing us to accurately tell if the resident has encountered a fall.
How we built it
We used a Particle Photon microcontroller alongside a MPU6050 gyro/accelerometer to be able to collect information about the movement of a residents foot and determine if the movement mimics the patterns of a typical fall. Once a typical fall has been read by the accelerometer, we used Twilio's RESTful API to transmit a text message to an emergency contact (or possibly a nurse/nurse station) so that they can assist the resident.
Challenges we ran into
Upon developing the algorithm to determine whether a resident has fallen, we discovered that there are many cases where a resident's feet could be in a position that can be interpreted as "fallen". For example, lounge chairs would position the feet as if the resident is laying down, so we needed to account for cases like this so that our system would not send an alert to the emergency contact just because the resident wanted to relax. To account for this, we analyzed the jerk (the rate of change of acceleration) to determine patterns in feet movement that are consistent in a fall. The two main patterns we focused on were:
- A sudden impact, followed by the shoe changing orientation to a relatively horizontal position to a position perpendicular to the ground. (Critical alert sent to emergency contact).
- A non-sudden change of shoe orientation to a position perpendicular to the ground, followed by a constant, sharp movement of the feet for at least 3 seconds (think of a slow fall, followed by a struggle on the ground). (Warning alert sent to emergency contact).
Accomplishments that we're proud of
We are proud of accomplishing the development of an algorithm that consistently is able to communicate to an emergency contact about the safety of a resident. Additionally, fitting the hardware available to us into the sole of a shoe was quite difficult, and we are proud of being able to fit each component in the small area cut out of the sole.
What we learned
We learned how to use RESTful API's, as well as how to use the Particle Photon to connect to the internet. Lastly, we learned that critical problem breakdowns are crucial in the developent process.
What's next for VATS
Next steps would be to optimize our circuits by using the equivalent components but in a much smaller form. By doing this, we would be able to decrease the footprint (pun intended) of our design within a clients shoe. Additionally, we would explore other areas we could store our system inside of a shoe (such as the tongue).