Icon Legend

This session is not in your schedule.

This session is in your schedule. Click again to remove it.

Orthopedic and Rehabilitation Engineering

Orthopedic and Rehabilitation Engineering - Poster Session D

Poster K11 - Real-time 3D Arm Motion Tracking Software Using 9 Degree of Freedom Inertial Measurement Units

Friday, October 25, 2024

3:30 PM - 4:30 PM EST

Location: Exhibit Hall E, F & G

Presenting Author(s)

AB

Abigail Beman (she/her/hers)

Student
University of Texas at Austin
Cedar Park, Texas, United States

Co-Author(s)

JM

Jason McDowell

Research Assistant
University of Tennesse at Chattanooga, United States

Presenting Author(s)

EK

Erkan Kaplanoglu

Professor
University of Tennesse at Chattanooga, United States

Introduction: The use of Inertial Measurement Units (IMUs) for motion tracking and visualization has become increasingly prevalent due to their compact size, affordability, and versatility. IMUs measure linear acceleration, rotational motion, and magnetic fields, making them ideal for various applications such as navigation, gaming, and healthcare. However, integrating IMU data into real-time 3D visualization software remains a challenge, particularly in ensuring accuracy and minimizing delays.
Many software options are available for musculoskeletal modeling and simulation, but they are often complex and frequently not optimized for real-time IMU data integration. In contrast, this project leverages Blender, an open-source 3D graphics software, to create a more accessible and user-friendly solution. Blender's robust Python scripting capabilities enable seamless integration of IMU data, providing a customizable and efficient platform for real-time 3D visualization.
This project presents the development of a software system for real-time 3D visualization of arm movements using three 9-degree-of-freedom IMUs. The primary objective is to create an accessible and effective tool for modeling and visualizing arm motion in real time, providing immediate feedback and accurate motion representation. Additionally, this system has potential future applications in rehabilitation. For instance, the development of a hand exoskeleton based on this project could offer detailed and interactive means of tracking and assisting arm and hand movements, enhancing patient outcomes and therapy effectiveness.

Materials and

Methods: In this study, our hardware setup includes three IMUs [1], each with an accelerometer, gyroscope, and magnetometer. These IMUs were placed on the upper arm, forearm, and palm to capture arm motion data. Blender, an open-source 3D graphics software with Python scripting capabilities, was used for real-time visualization. The IMUs connect wirelessly via Bluetooth using the Bleak library for Python, outputting data at 200 Hz for accurate motion tracking.
The data processing pipeline begins with initializing connections with the IMUs and reading raw IMU data. Orientation data is received as Euler angles (roll, pitch, yaw) and then conversion to quaternions to show more accurate and stable motion.
The system adjusts for initial offsets and calculates the relative positions of the IMUs. Each IMU's orientation is used to update the corresponding segment of a 3D arm model in Blender, ensuring realistic motion representation. Threading and efficient data handling techniques minimize delays, ensuring responsive visualization.
Drift calculations were performed to quantify the long-term stability and accuracy of the IMUs. Significant drift indicates potential issues with sensor calibration or environmental factors affecting measurements. By analyzing drift, we verified the reliability of the IMUs, ensuring that the visualization accurately reflects actual arm movements over extended periods.

Results, Conclusions, and Discussions: The software successfully connects to, configures, and reads data from three IMUs simultaneously. It provides real-time 3D visualization of arm movements, demonstrating the potential for accurate and responsive motion tracking. The system allows users to visualize the motion of an arm with three IMUs and collect data for future analysis. This visualization offers immediate feedback on arm movements, essential for various applications such as rehabilitation and interactive systems.
The project demonstrates the effectiveness of using IMUs for real-time 3D arm motion tracking and visualization. Integration with Blender allows for detailed and customizable visualizations, making it a versatile tool for real-time applications. The software's ability to provide real-time feedback and accurate motion representation highlights its potential for use in areas requiring precise motion tracking.
Future work will focus on validating the 3D arm visualization with ground-truth data from motion-capture systems and utilizing machine learning methods to predict hand and arm movements from IMU data. In particular, this project has the potential to leverage artificial intelligence to predict hand gestures based on arm movements, which could help patients with limited hand mobility.
By integrating IMU data into a customizable real-time 3D visualization, this project has successfully developed a modular and user-friendly system, offering significant implications for future research and applications in motion analysis and assistive technologies.

Acknowledgements (Optional): We thank the support of NSF REU #2149956 and the internal support from the Department of Computer Science and Engineering and the Office of the Vice Chancellor of Research at the University of Tennessee at Chattanooga.