Date of Award

Fall 2026

Abstract

Accurate gait analysis is important for clinical assessment, rehabilitation monitoring, fall-risk evaluation, sports performance analysis, and the control of assistive devices. While wearable sensor systems offer practical solutions for estimating key gait parameters outside the laboratory, their accuracy is often affected by sensor noise, drift, activity-dependent movement patterns, and changes in sensor response over time. These challenges limit their practical use in real-world settings, where continuous ground-truth validation and repeated laboratory calibration are not always available.This study addresses these challenges by developing context-aware wearable sensing methods for reliable gait analysis during everyday activities. The proposed approaches use information about the user’s activity, environmental conditions, gait events, inter-limb coordination, and sensor response changes to improve the estimation of clinically and biomechanically relevant gait parameters. First, activity recognition methods are introduced to identify motion intent across different terrains and to detect abnormal gait patterns such as foot shuffling. Building on this contextual information, this study then presents improved methods for estimating vertical foot trajectory using biomechanical constraints and for estimating stride length using contralateral foot motion data. Finally, this study introduces an auto-calibration method for long-term ground reaction force estimation using piezoresistive smart insoles. The method continuously updates the insole calibration during walking, enabling more accurate force estimation across daily activities while reducing the need for repeated laboratory recalibration. Together, these contributions advance the development of wearable gait analysis systems that are more accurate, adaptive, and practical for real-world health monitoring, rehabilitation, and assistive technology applications.

Document Type

Dissertation

First Advisor

Diliang Chen

Second Advisor

Andrew Kun

Third Advisor

Sanghee Moon

Department or Program

Electrical and Computer Engineering

Degree Name

Doctor of Philosophy

Download

Share

COinS