Personalizing gait phase and speed estimation from wearable sensors
Sparse Calibration-Based Personalization of Kernel-Based Gait Phase and Speed Estimation Using Wearable IMUs
Robotics
Summary
Measuring how people walk with wearable sensors can help with health and movement tracking, but everyone walks a little differently. The authors created a way to personalize walking measurements based on a few short calibration steps for each person. This makes the estimates of walking phases and speeds more accurate than just using average data from many people. They tested their method offline and in a small real-time pilot, showing it works on wearable devices but needs more testing with more people.
What this means in practice
- •For wearable device engineers: Create personalized gait monitoring software for fitness or rehabilitation devices by calibrating kernels with a few user walking trials.
- •For physical therapy technology teams: Improve accuracy of patient walking analysis tools by integrating sparse calibration to adapt gait phase and speed measurements to individuals.
Tested on one dataset.
Authors
Myeongju Cha, Pilwon Hur
Abstract
This paper presents sparse calibration-based personalization for kernel-based gait phase and walking speed co-estimation from wearable inertial measurement units (IMUs). A 28-speed reference library was constructed from bilateral thigh and shank trajectories in a 20-subject locomotion dataset. Rather than directly applying population-average kernels, the method combines a user-specific baseline estimated from three calibration speeds with a principal-component (PC) model of baseline-centered kinematic deviations. Offline validation showed that personalized kernels reduced phase error relative to the population kernel. In a single-participant online pilot evaluation, the personalized kernels produced numerically lower mean phase and speed errors and ran in real time on embedded hardware. However, the speed effect was not significant, and pairwise phase differences did not remain significant after multiple-comparison correction. The pilot evaluation establishes wearable implementation feasibility while indicating reference-to-online dataset mismatch and the need for larger-cohort validation.