EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Explainable Machine Learning for Human Activity Recognition Using Auxetic cTPU Knee-Worn Sensors

Abeer Elkhouly iD, Umar Asghar iD, Ganga Raj iD

Sensors · 2026

DOI: 10.3390/s26175548 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, 2026. Verfasst von Abeer Elkhouly und 2 weiteren Autor:innen. Beteiligte Einrichtung: School of Engineering, University of Wollongong in Dubai, Dubai 20183, United Arab Emirates. Hauptthemen laut MEDLINE: Wearable Electronic Devices, Machine Learning, Polyurethanes, Human Activities. Die Arbeit nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: University of Wollongong in Dubai.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper presents a wearable soft strain sensor based on a commercially available conductive thermoplastic polyurethane (cTPU) 3D-printed as an auxetic soft metamaterial for human activity recognition. The growing demand for flexible and wearable electronics, driven by advances in artificial intelligence, highlights the importance of such sensors in healthcare, medical rehabilitation, soft robotics, and human–machine interfaces. The auxetic cTPU sensor was mechanically and electrically characterized through empirical measurements and validated against numerical simulations. A single sensor mounted on a knee brace was used to collect gait signals across four activities: running, walking, standing, and sitting. Two classification approaches were investigated. A Long Short-Term Memory (LSTM) network was trained directly on the raw time-series signal, with the best configuration achieving 96% accuracy using Relative Standard Deviation Normalization with 50 hidden units. Traditional machine learning models, namely Random Forest and XGBoost, were trained on 30 extracted time-domain and frequency-domain features per motion cycle, achieving 100% and 97.33% accuracy, respectively, under five-fold cross-validation. To enhance model transparency, explainability analysis using SHAP identified power spectral density and the first harmonic frequency as the most consistently influential features across both models, with dynamic activities driven by frequency characteristics and stationary activities distinguished by signal mean amplitude. The results demonstrate that auxetic cTPU soft strain sensors combined with machine learning and explainable artificial intelligence provide an accurate and interpretable solution for wearable human activity recognition, highlighting their potential for applications in robotics, healthcare, and human–robot interfaces.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Wearable Electronic DevicesMachine LearningPolyurethanesHuman ActivitiesKneeHumansLong Short Term MemoryGaitSoft ComputingRoboticsWalking

Stichwörter der Autor:innen

auxetic materialsexplainable AI (XAI)human activity recognition (HAR)human gait classificationhuman–machine systemsintelligent sensorswearable devices

Substanzen

Polyurethanes

Autor:innen und Institutionen

  • School of Engineering, University of Wollongong in Dubai, Dubai 20183, United Arab Emirates.

ORCID-Profile: Abeer Elkhouly, Umar Asghar, Ganga Raj

Förderung

  • University of Wollongong in Dubai

Zitieren

APA

Elkhouly, A., Asghar, U., & Raj, G. (2026). Explainable Machine Learning for Human Activity Recognition Using Auxetic cTPU Knee-Worn Sensors. Sensors. https://doi.org/10.3390/s26175548

Vancouver

Elkhouly A, Asghar U, Raj G. Explainable Machine Learning for Human Activity Recognition Using Auxetic cTPU Knee-Worn Sensors. Sensors. 2026. doi:10.3390/s26175548

BibTeX
@article{Elkhouly2026,
  author = {Elkhouly, Abeer and Asghar, Umar and Raj, Ganga},
  title = {Explainable Machine Learning for Human Activity Recognition Using Auxetic cTPU Knee-Worn Sensors},
  journal = {Sensors},
  year = {2026},
  doi = {10.3390/s26175548}
}
Kostenlos starten →