Real-Time Motion Tracking and Classification Using Quaternion Sensor Data
University
Shawnee State University
Major
Computer Engineering Technology/Digital Sim.+Game Prog.
Student Type
Undergraduate Student
Presentation Types
Poster Group Presentation (Live)
Keywords:
Quaternions, Motion Classification, Wearable Sensors, Sports Analytics
Abstract
This project explores the use of wearable inertial sensors to classify human motion. Initial testing focuses on controlled motion trials, where users perform predefined actions while data is recorded and exported for analysis. Quaternion representations enable more robust tracking of rotational motion compared to traditional methods, making them well-suited for applications such as athletic performance analysis. Our long-term goal is to extend this system to classify complex movements in baseball, such as pitching and swinging, through refined sensor placement and data processing techniques.
Human and Animal Subjects
yes
IRB or IACUC Approval
yes
Faculty Mentor Name
Bob Newland
Faculty Mentor Title
Professor
Faculty Mentor Department
Engineering Technologies
Recommended Citation
Matthews, Hudson and Harris, Franklin, "Real-Time Motion Tracking and Classification Using Quaternion Sensor Data" (2026). Celebration of Scholarship. 4.
https://digitalcommons.shawnee.edu/cos/2026/PosterSession/4
Real-Time Motion Tracking and Classification Using Quaternion Sensor Data
This project explores the use of wearable inertial sensors to classify human motion. Initial testing focuses on controlled motion trials, where users perform predefined actions while data is recorded and exported for analysis. Quaternion representations enable more robust tracking of rotational motion compared to traditional methods, making them well-suited for applications such as athletic performance analysis. Our long-term goal is to extend this system to classify complex movements in baseball, such as pitching and swinging, through refined sensor placement and data processing techniques.