Next Article in Journal
Monitoring of Corroded and Loosened Bolts in Steel Structures via Deep Learning and Hough Transforms
Next Article in Special Issue
Interface Pressure System to Compare the Functional Performance of Prosthetic Sockets during the Gait in People with Trans-Tibial Amputation
Previous Article in Journal
Polarization-Insensitive Waveguide Schottky Photodetectors Based on Mode Hybridization Effects in Asymmetric Plasmonic Waveguides
Previous Article in Special Issue
Wearable Biofeedback Improves Human-Robot Compliance during Ankle-Foot Exoskeleton-Assisted Gait Training: A Pre-Post Controlled Study in Healthy Participants
Article

Body-Worn IMU Human Skeletal Pose Estimation Using a Factor Graph-Based Optimization Framework

1
Department of Aeronautics and Astronautics, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA 02139, USA
2
Industrial and Operations Engineering, Robotics Institute, University of Michigan, 1205 Beal Avenue, Ann Arbor, MI 48109, USA
*
Author to whom correspondence should be addressed.
Sensors 2020, 20(23), 6887; https://doi.org/10.3390/s20236887
Received: 20 October 2020 / Revised: 17 November 2020 / Accepted: 28 November 2020 / Published: 2 December 2020
(This article belongs to the Special Issue Smart Sensors: Applications and Advances in Human Motion Analysis)
Traditionally, inertial measurement units- (IMU) based human joint angle estimation requires a priori knowledge about sensor alignment or specific calibration motions. Furthermore, magnetometer measurements can become unreliable indoors. Without magnetometers, however, IMUs lack a heading reference, which leads to unobservability issues. This paper proposes a magnetometer-free estimation method, which provides desirable observability qualities under joint kinematics that sufficiently excite the lower body degrees of freedom. The proposed lower body model expands on the current self-calibrating human-IMU estimation literature and demonstrates a novel knee hinge model, the inclusion of segment length anthropometry, segment cross-leg length discrepancy, and the relationship between the knee axis and femur/tibia segment. The maximum a posteriori problem is formulated as a factor graph and inference is performed via post-hoc, on-manifold global optimization. The method is evaluated (N = 12) for a prescribed human motion profile task. Accuracy of derived knee flexion/extension angle (4.34 root mean square error (RMSE)) without magnetometers is similar to current state-of-the-art with magnetometer use. The developed framework can be expanded for modeling additional joints and constraints. View Full-Text
Keywords: inertial measurement system; human motion; joint angle; self-calibrating; biomechanics; optimization; factor graph; knee; soft tissue artifacts inertial measurement system; human motion; joint angle; self-calibrating; biomechanics; optimization; factor graph; knee; soft tissue artifacts
Show Figures

Figure 1

MDPI and ACS Style

McGrath, T.; Stirling, L. Body-Worn IMU Human Skeletal Pose Estimation Using a Factor Graph-Based Optimization Framework. Sensors 2020, 20, 6887. https://doi.org/10.3390/s20236887

AMA Style

McGrath T, Stirling L. Body-Worn IMU Human Skeletal Pose Estimation Using a Factor Graph-Based Optimization Framework. Sensors. 2020; 20(23):6887. https://doi.org/10.3390/s20236887

Chicago/Turabian Style

McGrath, Timothy, and Leia Stirling. 2020. "Body-Worn IMU Human Skeletal Pose Estimation Using a Factor Graph-Based Optimization Framework" Sensors 20, no. 23: 6887. https://doi.org/10.3390/s20236887

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop