Date of Award

Fall 12-18-2020

Level of Access Assigned by Author

Open-Access Thesis

Degree Name

Master of Science in Mechanical Engineering (MSME)

Department

Mechanical Engineering

Advisor

Babak Hejrati

Second Committee Member

Mohsen Shahinpoor

Third Committee Member

Vincent Caccese

Abstract

The detection of abnormal gait patterns is imperative for early diagnoses and treatment of serious health issues such as loss of mobility and balance, risk of falls, cardiovascular disease, etc. Human gait analysis is the tool used to detect abnormal gait patterns. Gait analysis also can help to understand the cause of gait abnormalities and to make a treatment plan for individual patients. Ground reaction forces (GRFs) at the foot during walking or running are essential for kinetic analysis of the human gait. However, there are no wearable systems that can directly measure three-directional (3D) forces during daily walking. Currently, the 3D GRFs are either measured by estimating the shear forces from the normal force measured by pressure insoles, or using instrumented treadmills, or walking on a limited number of force platforms in gait labs. There are always errors associated with estimating other force components from the normal one. Also, instrumented treadmills and force platforms are very expensive, and treadmill walking may not be sufficiently similar to over ground walking. So, a wearable system is needed to measure 3D ground reaction forces during walking for complete gait analysis in both indoor and outdoor conditions. The main challenge in developing such a wearable system or a smart shoe to measure 3D GRFs lies in the lack of a low-profile, lightweight, and portable force sensor that can measure 3D forces during the user’s walking. The main research objective here is to develop such a wearable force sensor for gait analysis applications. In this thesis work, the design, analysis, and fabrication of a capacitive-based 3D force sensor have been presented. The sensor mainly consists of an elastic element, three parallel plate capacitors, capacitance measuring electronics, and a bottom plate. The elastic component deforms under applied forces and this deformation leads to capacitance changes that are measured by the electronic circuit. An experimental setup was built to perform experiments needed to find the force-capacitance relationship and evaluate the sensor’s performance. A calibration matrix was found between the applied forces and the capacitance changes using the linear least-squares method. Five types of experiments were conducted to evaluate the developed sensor and the evaluated matrix, where the estimated forces were compared with the reference values obtained by commercially available force sensors. The developed sensor could measure forces in all three directions with the mean error of less than 4.5%. The experiments showed that the force sensor could successfully decouple the forces applied in different directions and had no residual offsets when no force was applied. Also, no drifts and changes in the behavior of the sensor were observed after a long period of usage. The developed sensor also demonstrated adequate repeatability, hysteresis characteristics, and dynamic response. The measurement errors increased in some scenarios, in which the forces were simultaneously applied in different directions. Future work can be done to reduce the error in the combined conditions.

Share