Date of Award

8-2026

Degree Name

MS in Electrical Engineering

Department/Program

Electrical Engineering

College

College of Engineering

Advisor

Sachiko Matsumoto

Advisor Department

Electrical Engineering

Advisor College

College of Engineering

Abstract

Active lower-limb prostheses use intent-recognition systems to identify a user’s locomotion mode and select an appropriate control strategy, but sensor configurations that perform well offline may be unsuitable for resource-constrained embedded hardware. Existing sensor-selection methods generally prioritize classification accuracy without directly accounting for processing latency, memory usage, or other hardware-dependent requirements. To address this limitation, this thesis develops a hardware-in-the-loop source-selection framework for embedded classification of level walking, ramp ascent, ramp descent, stair ascent, and stair descent using multimodal biomechanical data from transtibial amputee participants. Subject-specific linear support vector machine classifiers were evaluated using trial-held-out validation, and candidate configurations from predefined minimal and reduced sensor-source pools were tested on an STM32L476RG microcontroller. The complete offline configuration achieved an overall accuracy of 54.89%, while the hardware-in-the-loop experiment produced 54.24% accuracy with a subset of sources, completed all classifications within the required 50-ms update interval, and remained within the configured memory limits. These findings demonstrate that hardware characteristics not captured through offline evaluation alone can be incorporated into sensor-source selection and establish hardware-in-the-loop testing as a practical step toward fully integrated intent-recognition systems for active lower-limb prostheses.

Download

Share

COinS