Portable Gait Lab: Estimating 3D GRF Using a Pelvis IMU in a Foot IMU Defined Frame

Research output: Contribution to journalArticleAcademicpeer-review

33 Citations (Scopus)
180 Downloads (Pure)

Abstract

Ground Reaction Forces (GRF) during gait are measured using expensive laboratory setups such as in-floor or treadmill force plates. Ambulatory measurement of GRF using wearables enables remote monitoring of gait and balance. Here, we propose using an Inertial Measurement Unit (IMU) mounted on the pelvis to estimate the GRF during gait in daily life. Calibration procedures and an Error State Extended Kalman filter (EEKF) were used to transform the accelerations at the center of mass (CoM) to the 3D GRF. The instantaneous 3D GRF was estimated for different overground walking patterns and compared with the 3D GRF measured using the reference ForceShoe™ system. Furthermore, we introduce a changing reference frame called the current step frame that followed the direction of each step made. The frame was defined using movement of the feet, and the estimated GRF were expressed in this new frame. This allowed direct comparison and validation with the reference. The mean and standard deviation of error between the estimated instantaneous 3D GRF and the reference, normalized against the range of the reference, was 12.1 ± 3.3% across all walking tasks, in the horizontal plane. The error margins show that a single pelvis IMU could be a minimal and ambulatory sensing alternative for estimating the instantaneous 3D components of GRF during overground gait.
Original languageEnglish
Article number9072192
Pages (from-to)1308-1316
Number of pages9
JournalIEEE transactions on neural systems and rehabilitation engineering
Volume28
Issue number6
DOIs
Publication statusPublished - 20 Apr 2020

Keywords

  • Ambulation
  • Ground reaction forces
  • Inertial measurement unit
  • Minimal sensing
  • Gait

Fingerprint

Dive into the research topics of 'Portable Gait Lab: Estimating 3D GRF Using a Pelvis IMU in a Foot IMU Defined Frame'. Together they form a unique fingerprint.

Cite this