Thesis work: Studying the influence of a fabric substrate on wearable stretch sensors
Purpose:
Explore the (typical and over-simplified!) design trade-off between comfort and accuracy of soft wearable sensors
Build the case that fabric-based substrates can add complexity in sensor performance, especially due to its inherent mechanical properties when stretched or manipulated on a user's body.
Create set of experiments to investigate How much influence or "noise" is the fabric contributes to a particular stretch sensor
Based on assumption sensor reading = amount of stretch + fabric substrate influence or bias
Test design choices to quantify type and magnitude of influence on sensor performance
Approach:
Background research into wearable sensing space, specifically strain/stretch sensing
Develop key research questions and design experiments per scientific methodology
Relate some tabletop testing to real situations via data modeling (single sensor - linear regression, multiple sensor samples, machine learning models (decision trees and neural network algorithms via Matlab))
Fabricate soft sensor prototypes and required electronic circuit boards for testing
Test & analyze findings.
Present work in a public forum (thesis defense)
Project duration: 1.5 years
Role & key contributions:
Primary researcher
Defend thesis in a public presentation
Final product overview: The thesis was completed on time and without major edits!