Garment-based measurement of human upper body orientation using miniature inertial sensing units

Type: Master thesis
Student: Rene Winkler
Advisor: Holger Harms, Johannes Schumm

Measurement of human upper body motions is beneficial in miscellaneous applications like virtual reality, movement science, robotics and other human-machine interactions. However, state-of-the-art systems are cumbersome, obtrusive and do not allow measurements in daily life.

This master thesis addressed the issue concerning the design and the implementation of a garment-based human upper body orientation measurement system adapted from inertial and magnetic sensors. During its development, careful attention was paid to its miniaturization and its weight enabling an unobtrusive and comfortable integration in an everyday shirt.

The developed system consists of a central master device and five sensing units. The sensing units are attached on every upper body segment providing its orientation. They contain a set of three sensor modalities measuring the acceleration, the magnetic field and the angular velocity. This data is processed by appropriate signal fusion algorithms for real-time human motion tracking.

A couple of meaningful hardware characterization tests were performed in order to verify the proper functioning of the system. Validation tests for checking the implemented signal fusion algorithms highlighted their effectiveness: A static orientation accuracy of 2° and a dynamic orientation accuracy of 12° were obtained.

Several demonstrators comprising a Java application, a stand-alone orientation module and an animated Matlab 3D stickman completed this master thesis.