Gait Analysis Project

A gait is a person's pattern of walking. This project was about gait analysis, which is the systematic study of human motion with the goal of assisting those with conditions that affect their ability to walk, such as helping athletes move more effectively and identifying posture or movement related issues. Using visual observations and movement measurements, we wanted to answer the question: How can we predict someone's weight based on their average acceleration?

Through data analysis, mathematical modeling, and reporting, we were able to produce our results for our gait analysis. We took several different measurements of group members and classmates to incorporate in our predictive model. Some of the things we measured were height, length of knee to toe, hip to hip, and their weight. Other measurements we incorporated were the length between the tip of your knee to your hip, as well as the length from hip to hip and shoe size (in mens). Next, we downloaded the app Physics Toolbox Accelerometer to track each of us walking a specified distance. We recorded the way our body shifted horizontally and vertically, as well as our acceleration (g-force).

Our next step was to input the data into a spreadsheet to analyze our findings. We were able to create multiple graphs to represent the relationship between our g-forces and time elapsed walking. Through comparing these relationships, we could work towards creating our predictive model.

To predict one’s approximate weight with our model, we need:

1. Your average acceleration (g-force)

The first step in the process is to find the subject's average acceleration. It should be around 1. To find this, you need to fix a phone downloaded with a software onto one's back and have them walk a set distance. The software we used was called “Physics Toolbox accelerometer.” The next step is to put all of your data into a spreadsheet. This will make it easy to calculate the average and visualize your data in graphs. The final step was to see where your number falls on the table below and see how accurate our prediction is.

Average acceleration Predicted weight range

1.0118-1.0111 110-120 lbs

1.0111-1.0107 120-130 lbs

1.0107-1.0104 130-140 lbs

1.0104-1.0102 150-160 lbs

1.0102-1.0099 160-170 lbs

1.0099- 1.0025 170-180 lbs

1.0095-1.0091 180-190 lbs

Here was our data in the model:

Name Actual weight Avg Acceleration Predicted weight

Josh M 190 1.0091 180-190

Sean J 171 1.0099 170-180

Kiana S 135 1.0122 N/A (outlier)

Ben D 112 1.0115 110-120

Ally B (test) 115 1.0111 110-120

A pattern we saw was that the heavier you get, the slower your average acceleration. We predict that this is because as you get heavier, it takes more energy to move, and thus you might inadvertently walk slower.

Overall the project was a success, but we did encounter struggles throughout our gait analysis process. Initially, our initial formula took into account the variable we were predicting. As it incorporated weight, it was obvious that this would not be a possible model for predicting other group members' weights. It was a continued struggle to find patterns among the data to create our predictive model. We spent many hours experimenting and trying out different calculations. We took a step back and decided to use a simple table instead of a lengthy equation for our model. While this strategy might have reduced the accuracy of our model slightly, we are confident that finding another way would have taken an unnecessary and obscene amount of time. This project was a great lesson in perseverance and dividing up work evenly. It took the entire team to compartmentalize the report, analysis, and presentation. We are very proud of our achievements and what we have learned.

Something I feel that I did well in this project was communicating. My group collaborated very well and did a great job exchanging ideas for how to accomplish various aspects of this project. We were all able to contribute to how we wanted to test and take data, and eventually produce a model that reflected our hard work. Another thing I believe I did well was problem solving and adapting to challenges we faced throughout the entirety of the project. We did encounter many obstacles when attempting to finalize a predictive model that worked, but my group members and I worked very hard to continue testing new formulas until we were successful.

I feel that I could have improved on my empathy and patience in this project. As it had many complicated components and took a very long time, I found myself often getting frustrated. When a potential predictive model did not work or when we had graphing struggles, it led to feelings of discouragement. Although it all worked out in the end, I think I can do a better job in the future of staying calm. I also think I could have done better with personal organization. While my group divided up work, I sometimes found myself struggling to manage my own load. I let a combination of outside work/life stress me out. I got everything done, but feel that I could have managed myself better.

Overall, this project led me to learn much more about engineering analysis and data collecting. Although it was complicated and very challenging at times, I can say with confidence my knowledge has expanded and we produced a successful model after hard work and persistence. My group and I are now more confident in our ability to synthesize lots of data and collaborate as a team in a goal to make the best Capstone project possible.