Research Repository

Lower Limb Kinematics Trajectory Prediction Using Long Short-Term Memory Neural Networks

Zaroug, A, Lai, Tze Huel ORCID: 0000-0003-3459-7709, Mudie, Kurt ORCID: 0000-0001-9328-4759 and Begg, Rezaul ORCID: 0000-0002-3195-8591 (2020) Lower Limb Kinematics Trajectory Prediction Using Long Short-Term Memory Neural Networks. Frontiers in Bioengineering and Biotechnology, 8. ISSN 2296-4185

[img]
Preview
Text
Lower Limb Kinematics Trajectory Prediction Using Long Short-Term Memory Neural Networks.pdf - Published Version
Available under License Creative Commons Attribution.

Download (6MB) | Preview

Abstract

© Copyright © 2020 Zaroug, Lai, Mudie and Begg. This study determined whether the kinematics of lower limb trajectories during walking could be extrapolated using long short-term memory (LSTM) neural networks. It was hypothesised that LSTM auto encoders could reliably forecast multiple time-step trajectories of the lower limb kinematics, specifically linear acceleration (LA) and angular velocity (AV). Using 3D motion capture, lower limb position–time coordinates were sampled (100 Hz) from six male participants (age 22 ± 2 years, height 1.77 ± 0.02 m, body mass 82 ± 4 kg) who walked for 10 min at 5 km/h on a 0% gradient motor-driven treadmill. These data were fed into an LSTM model with a sliding window of four kinematic variables with 25 samples or time steps: LA and AV for thigh and shank. The LSTM was tested to forecast five samples (i.e., time steps) of the four kinematic input variables. To attain generalisation, the model was trained on a dataset of 2,665 strides from five participants and evaluated on a test set of 1 stride from a sixth participant. The LSTM model learned the lower limb kinematic trajectories using the training samples and tested for generalisation across participants. The forecasting horizon suggested higher model reliability in predicting earlier future trajectories. The mean absolute error (MAE) was evaluated on each variable across the single tested stride, and for the five-sample forecast, it obtained 0.047 m/s2 thigh LA, 0.047 m/s2 shank LA, 0.028 deg/s thigh AV and 0.024 deg/s shank AV. All predicted trajectories were highly correlated with the measured trajectories, with correlation coefficients greater than 0.98. The motion prediction model may have a wide range of applications, such as mitigating the risk of falls or balance loss and improving the human–machine interface for wearable assistive devices.

Item Type: Article
Uncontrolled Keywords: short-term memory, neural networks, multiple layer perceptrons, temporal correlations, human activity recognition, steady-state walking, human movement prediction
Subjects: Current > FOR Classification > 1702 Cognitive Science
Current > Division/Research > College of Science and Engineering
Current > Division/Research > Institute for Health and Sport
SWORD Depositor: Symplectic Elements
Depositing User: Symplectic Elements
Date Deposited: 20 Oct 2020 04:39
Last Modified: 20 Oct 2020 04:39
URI: http://vuir.vu.edu.au/id/eprint/41422
DOI: https://doi.org/10.3389/fbioe.2020.00362
ePrint Statistics: View download statistics for this item
Citations in Scopus: 0 - View on Scopus

Repository staff only

View Item View Item

Search Google Scholar