Inspiration
One of our team member's roommate frequently has knee problems, which, of course, affect his overall wellness. As we looked more into it, we discovered that knee problems are incredibly common—knee surgery is performed about 800,000 times per year in the US alone. And physical therapy, while essential, can only take place for small, constrained windows of time. Robotic assistants, while extensive, are extremely expensive. We decided to make a cheap device that can provide assistance or resistance, allowing people recovering from surgery to practice these motions on their own and track their progress, helping motivate them and helping their physical therapist understand their recovery trajectory.
What it does
Our device straps onto people's knees and dynamically adapts to their attempts at motion by providing assistance or resistance based on their desired setting through a mobile app. It then tracks the amount of assistance they needed, or the amount of resistance they were able to face, and graphs their progress over time for both them and their physical therapist to see the way they are progressing.
How we built it
We 3D printed physical braces for the top and bottom of people's thigh and shins, and then wired a Raspberry Pi Pico to a set of imus, motors, and a Bluetooth transceiver. A web application captures the Bluetooth data and advertises it to a Swift iOS app. This app uses SwiftUI to display dynamic graphs to the user that demonstrate their progress, and allows the user to switch between assistance and resistance modes depending on their point in recovery.
Challenges we ran into
About ten minutes ago, we discovered that our microcontroller was fried. We also discovered that our Bluetooth module was incompatible with our iPhone application, hence the need for the Flask web application. Many of our 3D prints failed. Also, due to our inexperience with Swift, we have thus far been unable to figure out how to make our graph dynamically update. Hence, this project is still very much a work-in-progress.
Accomplishments that we're proud of
We successfully programmed a motor controller using a PWM controller to maintain a target velocity despite the variable amount of support coming from the human user, calculate the force coming from the user and coming from the motor, and stop the assistance/resistance when no motion occurs to make sure that we do not harm the user. We were able to feed this data to a web application and read it into an iOS app. Despite many 3D prints failing, we were also able to design extensive models of how this design might work and print some components that we can display now.
What we learned
We gained more familiarity with Swift and iOS app development. We also learned the importance of planning our software and firmware stack in advance, so that we can avoid issues with compatibility. Regarding the theme of wellness, the hackathon also served as an important reminder of the importance of proper sleep habits for overall wellness.
What's next for Super Knee
This project is very much still in-progress, so our immediate next steps are to successfully make the iOS graph dynamically update and put the final touches on it so that it provides a user-friendly experience, as well as replacing the microcontroller and connecting the parts to make a properly combined unit. In the further future, we believe that the technology behind Super Knee has many applications to the general field of physical therapy, and could even be applied to ideas like helping the general public with difficult yoga poses that are known to be beneficial for mental wellness.
Log in or sign up for Devpost to join the conversation.