This thesis presents the development towards a system that can capture and quantify motion for applications in biomechanical and medical fields demanding precision motion tracking using the lighthouse technology. Commercially known as SteamVR tracking, the lighthouse technology is a motion tracking system developed for virtual reality applications that makes use of patterned infrared light sources to highlight trackers (objects embedded with photodiodes) to obtain their pose or spatial position and orientation. Current motion capture systems such as the camera-based motion capture are expensive and not readily available outside of research labs. This thesis provides a case for low-cost motion capture systems. The technology is applied to quantify motion to draw inferences about biomechanics capture and analysis, quantification of gait, and prosthetic alignment. Possible shortcomings for data acquisition using this system for the stated applications have been addressed. The repeatability of the system has been established by determining the standard deviation error for multiple trials based on a motion trajectory using a seven degree-of-freedom robot arm. The accuracy testing for the system is based on cross-validation between the lighthouse technology data and transformations derived using joint angles by developing a forward kinematics model for the robot’s end-effector pose. The underlying principle for motion capture using this system is that multiple trackers placed on limb segments allow to record the position and orientation of the segments in relation to a set global frame. Joint angles between the segments can then be calculated from the recorded positions and orientations of each tracker using inverse kinematics. In this work, inverse kinematics for rigid bodies was based on calculating homogeneous transforms to the individual trackers in the model’s reference frame to find the respective Euler angles as well as using the analytical approach to solve for joint variables in terms of known geometric parameters. This work was carried out on a phantom prosthetic limb. A custom application-specific motion tracker was also developed using a hardware development kit which would be further optimized for subsequent studies involving biomechanics motion capture.
| Identifer | oai:union.ndltd.org:UMASS/oai:scholarworks.umass.edu:masters_theses_2-1735 |
| Date | 25 October 2018 |
| Creators | Sitole, Soumitra |
| Publisher | ScholarWorks@UMass Amherst |
| Source Sets | University of Massachusetts, Amherst |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Masters Theses |
Page generated in 0.0028 seconds