Date of Award

6-3-2026

Document Type

Thesis

Publisher

Santa Clara : Santa Clara University, 2026

Departments

Computer Engineering; Computer Science and Engineering; Electrical Engineering; Electrical and Computer Engineering

First Advisor

Michael Schimpf

Second Advisor

Maria Kyrarini

Abstract

Traditional walkers provide essential physical support for people recovering from injury, surgery, or reduced mobility, but do not collect objective information about how a person walks. This thesis presents the design, implementation, and evaluation of a SmartWalker system that transforms a standard mobility walker into a portable gait data collection platform. The system integrates Intel RealSense cameras mounted on a walker, NVIDIA Jetson embedded computing units, local storage, and a graphical user interface to record lower-body movement, process video locally, and present clinician-reviewable gait metrics.

The thesis describes the motivation for the monitoring of gait based on walking, the user research and the needs of stakeholders, the design of the physical and computational system, and the experimental evaluation of the prototype. The recorded walking videos are processed using Sports2D to extract lower-body pose and joint-angle data, then analyzed with a Fast Fourier Transform to estimate the walking frequency. The prototype was evaluated through 10- meter walking trials at 80 and 100 bpm, with and without a restrictive medical boot to simulate gait changes. The results show that the system can collect usable walking data and provide objective frequency-based gait information, while also addressing practical constraints, engineering standards, privacy, safety, usability, and ethical considerations.

Download

Share

COinS