Waddle

Device not on leg
Device not on leg
Device not on leg
Device on leg
Device on leg

Inspiration

It was the 47th minute of what had been a grueling soccer game against OU where everything changed. Harnold, UT’s star player was sprinting towards a loose ball near the sideline when a kick from the opposing defender sent the ball directly into his left shin. He tried to shake it off like usual, but this time Harnold waddled off the field in pain. When my team saw this, we knew we had to aid in his recovery. We searched for a solution on the internet, however we found nothing to detect how well his recovery is going. Determined to get Harnold back on the field, we built Waddle in order to make sure he can make it to the next game.

What it does

Waddle is a wearable device that goes around the knee and tracks rotation during walking, standing, or exercising. It keeps track of knee flexion for both IMUs to calculate knee flexion. The ESP32 sends the data through wifi so it can be processed and a score given to the user and the data is stored in an SQL database, providing feedback on leg movement to help patients and therapists monitor recovery progress over time after injury, making rehab more measurable and engaging.

How we built it

Waddle’s prototype consists of 2 IMUs and an ESP32 connected via a breadboard, electrical tape, surgical tubing, and 3D printed parts. It keeps track of rotation for both IMUs to calculate knee flexion. The ESP32 then sends this data through wifi to a websocket in the FastAPI server. The data is processed to detect steps, which generates a list of max and min flexion for each step for each walking session. The score is then generated based on Range of Motion (max flexion - min flexion) and how close the user is to some target values. The current weights and targets: 60% is Avg ROM (Target: 43°) 15% is Max ROM: (Target: 55°) 25% is Standard Deviation: (Target: 0)

Challenges we ran into

Deducing what values received count as a single step, sensor filtering (we used complimentary filtering to reduce noisy data from the IMUs), and figuring out how to strap the device to the leg with the materials we had available. Also we had to learn about database management and front-end development.

Accomplishments that we're proud of

As it was the entire team’s first hackathon, we didn’t expect it coming into Hack Hack Goose, however, the team walked out of the hackathon with a working prototype. Our team is also proud that Waddle was built with a real goose in mind. Harnold isn’t just a user, he’s the reason Waddle exists. Knowing that these tools could one day help him and athletes like him get back on the field safely makes all the obstacles we pushed through worth it.

What we learned

The team’s personal development had a major focus on front-end development. Considering everyone’s background in computer science and electrical engineering, everyone’s skills were mainly based around back-end systems. Therefore, being able to work on front-end design helped level out the skills of the group. The team also learned about database management, specifically using SQL as a way to store and retrieve data to make comparisons with data collected over time. Learning sensor filtering and data processing was another skill the group gained during this project.

What's next for Waddle

Ideally the front-end for Waddle would instead be an app so that users can more easily download and use Waddle, and so connecting the device would be easier for the average person. Also, a 2nd set of trackers on the other leg would be better to compare injured leg to healthy leg, however we did not have access to more IMU’s at the time.

Possible features: -Ability to share data to doctors -Use an AI model to track specific user walking habits -Based on changes in walking habits, make recommendations for doctor visits -Ability to edit the weights and targets used to calculate session scores based on specific user needs. -User can label sessions as exercise sessions to track reps and ROM for specific workouts