Faculty member: Dr. Brad Bennett (Kinesiology)
Dr. Bennett is looking for students from the following disciplines: Computer Science, Statistics, Engineering, and Mathematics
Discipline: Biomechanics (the study of how the skeletal and musculature systems work under different conditions)
General description of Dr. Bennett’s expertise:
Dr. Bennett received his BS in Mechanical Engineering from the University of Wisconsin and his MS and PHD also in Mechanical Engineering from the High Temperature Gasdynamics Laboratory at Stanford University. After 10 years in aerospace research in computational fluid dynamics. Dr. Bennett switched his professional interest to the study of human movement.
Dr. Bennett's research interests include the study of human gait and using the biomechanics of movement as a window into the organization of human movement. He believes the best movement data is that which can be captured during daily living and is developing new techniques to measure gait outside of the lab.
The project work will include programming statistical concepts of machine learning and sophisticated integration of measured data. This research project will study how people walk and run in daily life, data is collected from Inertial Measurement Units (IMUs) attached to an individual’s feet. Raw data includes linear accelerations and angular velocity in all three dimensions. Unfortunately, this data is of low quality with lots of noise and drift. Assistance is needed to analyze this data. Issue: 1) Correctly filter data to remove noise, 2) Be able to identify when person is walking or running, and, 3) Be able to determine the stride length during walking and running.
Specifically, machine learning algorithms need to be refined or redesigned as necessary to identify periods of walking and running. Secondly, we need to develop double integration methods to determine gait speed and stride length to remove drift from data. Original versions of software are written in Matlab.
Student-researcher expectations and responsibilities: