EUVIMEDEUVIMED entdecken
Crossref

Single Accelerometer to Recognize Human Activities Using Neural Networks

Sai Siddarth Vakacherla, Prakyath Kantharaju, Meet Mevada, Myunghee Kim

Journal of Biomechanical Engineering · 2023

DOI: 10.1115/1.4056767 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Biomechanical Engineering, 2023. Verfasst von Sai Siddarth Vakacherla und 3 weiteren Autor:innen. Beteiligte Einrichtung: Rehabilitation Robotics Laboratory, Mechanical and Industrial Engineering, University of Illinois at Chicago, Chicago, IL 60607. Hauptthemen laut MEDLINE: Activities of Daily Living, Neural Networks, Computer. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 67 Literaturangaben. Gefördert von: U.S. Army Research Laboratory.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Exoskeletons have decreased physical effort and increased comfort in activities of daily living (ADL) such as walking, squatting, and running. However, this assistance is often activity specific and does not accommodate a wide variety of different activities. To overcome this limitation and increase the scope of exoskeleton application, an automatic human activity recognition (HAR) system is necessary. We developed two deep-learning models for HAR using one-dimensional-convolutional neural network (CNN) and a hybrid model using CNNs and long-short term memory (LSTM). We trained both models using the data collected from a single three-axis accelerometer placed on the chest of ten subjects. We were able to classify five different activities, standing, walking on level ground, walking on an incline, running, and squatting, with an accuracy of 98.1% and 97.8%, respectively. A two subject real-time validation trial was also conducted to validate the real-time applicability of the system. The real-time accuracy was measured at 96.6% and 97.2% for the CNN and the hybrid model, respectively. The high classification accuracy in the test and real-time evaluation suggests that a single sensor could distinguish human activities using machine-learning-based models.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Activities of Daily LivingNeural Networks, ComputerHumansMachine LearningHuman ActivitiesAccelerometry

Autor:innen und Institutionen

  • Rehabilitation Robotics Laboratory, Mechanical and Industrial Engineering, University of Illinois at Chicago, Chicago, IL 60607.

Förderung

  • U.S. Army Research Laboratory

Zitieren

APA

Vakacherla, S. S., Kantharaju, P., Mevada, M., & Kim, M. (2023). Single Accelerometer to Recognize Human Activities Using Neural Networks. Journal of Biomechanical Engineering. https://doi.org/10.1115/1.4056767

Vancouver

Vakacherla SS, Kantharaju P, Mevada M, Kim M. Single Accelerometer to Recognize Human Activities Using Neural Networks. Journal of Biomechanical Engineering. 2023. doi:10.1115/1.4056767

BibTeX
@article{Vakacherla2023,
  author = {Vakacherla, Sai Siddarth and Kantharaju, Prakyath and Mevada, Meet and Kim, Myunghee},
  title = {Single Accelerometer to Recognize Human Activities Using Neural Networks},
  journal = {Journal of Biomechanical Engineering},
  year = {2023},
  doi = {10.1115/1.4056767}
}
Kostenlos starten →