Gait Analysis
Project Description
A person's gait is typically defined as the way in which they move their limbs to get around, but it is also referred to as the way someone walks. In this project, we had to collect data on multiple subjects gaits and use that data to create a predictive model that could tell you something like height or age of an unknown subject. We used four different subjects and collected the time it took them to walk 10 steps, the distance they traveled, time elapsed, and the acceleration in all three axes using an app called Physics Toolbox Accelerometer. There were six total trials taken, two in three different spots: the left shin, right shin, and the lower abdomen, about where the center of gravity would be. This data was then compiled into a spreadsheet and further analyzed. We were tasked with creating a predictive model based on the data we collected that would accurately predict something about the person, like height, age, or leg length.To obtain our predictive model, a lot of trial and error was used. There were some odd patterns in the data, which made it hard to create a predictive model that was very accurate. For example, one of the subjects was shorter than another, yet that first subject had a longer stride length. This made the predictive model less accurate, because it was based on the assumption that the taller you were, the longer your stride length. We then had to create a report detailing our findings, our data, predictive model, and an analysis of the predictive model in both its creation and application. Finally, we created a mini presentation to demonstrate and explain our predictive model to the class. These can both be found below.
A person's gait is typically defined as the way in which they move their limbs to get around, but it is also referred to as the way someone walks. In this project, we had to collect data on multiple subjects gaits and use that data to create a predictive model that could tell you something like height or age of an unknown subject. We used four different subjects and collected the time it took them to walk 10 steps, the distance they traveled, time elapsed, and the acceleration in all three axes using an app called Physics Toolbox Accelerometer. There were six total trials taken, two in three different spots: the left shin, right shin, and the lower abdomen, about where the center of gravity would be. This data was then compiled into a spreadsheet and further analyzed. We were tasked with creating a predictive model based on the data we collected that would accurately predict something about the person, like height, age, or leg length.To obtain our predictive model, a lot of trial and error was used. There were some odd patterns in the data, which made it hard to create a predictive model that was very accurate. For example, one of the subjects was shorter than another, yet that first subject had a longer stride length. This made the predictive model less accurate, because it was based on the assumption that the taller you were, the longer your stride length. We then had to create a report detailing our findings, our data, predictive model, and an analysis of the predictive model in both its creation and application. Finally, we created a mini presentation to demonstrate and explain our predictive model to the class. These can both be found below.
Gait Analysis Report |
Gait Analysis Presentation |
Terms and Definitions
Gait Analysis: An in depth observation and analysis of the way people, or animals, move their limbs to move the rest of their body.
Gait: The manner in which a person moves their limbs to get around.
Predictive Model: An equation that uses input data to accurately predict an outcome.
Accelerometer: A device that is able to measure the acceleration of an object.
Gait Analysis: An in depth observation and analysis of the way people, or animals, move their limbs to move the rest of their body.
Gait: The manner in which a person moves their limbs to get around.
Predictive Model: An equation that uses input data to accurately predict an outcome.
Accelerometer: A device that is able to measure the acceleration of an object.
Reflection
This project was a rough one. There were two days to collect data and we only collected a little each day because we were trying to figure out what data we wanted to collect in order to best make the best and most accurate predictive model we could. After these two days, the project didn't really get worked on. Better communication outside of school could have solved this problem. This would have allowed us to collect the data we needed and make sure each person was doing their part in the project. One thing we did do well was collect quality data. The procedure was well planned, and having two trials helped make the data more accurate. This led to a better predictive model and ultimately a better project. Another area we needed to improve upon was the overall focus of our group. The two days we had to collect data were two distracted days, and that led to a lack of data being collected. This forced us to collect data outside of school, which could have been avoided had we been less distracted those two days. One other thing we did really well was that we were very thorough in our data collection and analysis. We made sure to factor everything into account before we created our predictive model. This project wasn't the best due to the amount of work that had to be done outside of class, but it was still very interesting to work on.
This project was a rough one. There were two days to collect data and we only collected a little each day because we were trying to figure out what data we wanted to collect in order to best make the best and most accurate predictive model we could. After these two days, the project didn't really get worked on. Better communication outside of school could have solved this problem. This would have allowed us to collect the data we needed and make sure each person was doing their part in the project. One thing we did do well was collect quality data. The procedure was well planned, and having two trials helped make the data more accurate. This led to a better predictive model and ultimately a better project. Another area we needed to improve upon was the overall focus of our group. The two days we had to collect data were two distracted days, and that led to a lack of data being collected. This forced us to collect data outside of school, which could have been avoided had we been less distracted those two days. One other thing we did really well was that we were very thorough in our data collection and analysis. We made sure to factor everything into account before we created our predictive model. This project wasn't the best due to the amount of work that had to be done outside of class, but it was still very interesting to work on.