Prediction of Three-Directional Ground Reaction Forces during Walking Using a Shoe Sole Sensor System and Machine Learning
Problem
Measuring three-directional ground reaction forces during walking normally requires lab force plates; a wearable shoe sole sensor system was developed to predict them.
Why it matters
Portable GRF measurement enables gait analysis outside labs, relevant to clinical movement assessment.
What the AI did
Regression models (multiple linear regression and Gaussian process regression) predicted three-directional ground reaction forces from four triaxial sole force sensors, validated against force plate measurements in eight participants.
Results
Percent RMSE below 15% for anteroposterior and vertical GRFs; mediolateral direction excepted
Impact
Wearable prediction of GRFs achieved stated error thresholds for two of three force directions.
Record
| Venue | Sensors |
|---|---|
| Published | 2023 |
| Work type | article |
| Authors | Takeshi Yamaguchi, Yuya Takahashi, Yoshihiro Sasaki |
| Citations | 20 (OpenAlex, at import time) |
| References | 36 |
| Model or method | Multiple linear regression and Gaussian process regression (GPR) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3390/s23218985 |
| Open access | gold |
| Keywords | ground reaction forces, wearable sensors, Gaussian process regression, gait analysis, force plate |
Importance rating
real but narrower advance
Portable GRF measurement enables gait analysis outside labs for clinical movement assessment; useful niche capability, no scale or population impact stated.