Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Ultra-High Gain Vertically Stacked WS<sub>2</sub> NMOS Inverter Enabled by Co-Optimization of Layer Number and Contact Metal.

ACS nano·2026
Same author

Clonal evolution and mutational trajectories of metastatic colorectal cancer shaped by anticancer therapies.

Nature communications·2026
Same author

Treatment Outcomes and Prognostic Factors of Metastasis-Directed Radiation Therapy for Oligometastatic Endometrial Cancer.

Advances in radiation oncology·2026
Same author

1 V-Operable High-Mobility Amorphous Indium-Tin-Gallium-Zinc Oxide Thin-Film Transistors with High-<i>k</i> Hafnium Dioxide Gate Dielectrics through Continuous Atomic Layer Deposition.

ACS applied materials & interfaces·2026
Same author

Topological resilience of EEG-based functional networks during virtual reality-induced emotional states.

Frontiers in human neuroscience·2026
Same author

Metastasis-directed radiotherapy for oligometastatic cervical carcinoma: Identifying potential beneficiaries.

Clinical and translational radiation oncology·2026

Related Experiment Video

Updated: Aug 16, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

4.0K

Real-Time Human Activity Recognition with IMU and Encoder Sensors in Wearable Exoskeleton Robot via Deep Learning

Ismael Espinoza Jaramillo1, Jin Gyun Jeong1, Patricio Rivera Lopez2

  • 1Department of Electronics and Information Convergence Engineering, Kyung Hee University, Yongin 17104, Republic of Korea.

Sensors (Basel, Switzerland)
|December 23, 2022
PubMed
Summary

This study developed a real-time activity recognition system for exoskeleton robots using deep learning models and sensor data. The system accurately identifies eight common human activities, enhancing robot control assistance.

Keywords:
deep learning networksencodersinertial measurement unitreal-time human activity recognitionwearable exoskeleton robot

More Related Videos

The Muscle Cuff Regenerative Peripheral Nerve Interface for the Amplification of Intact Peripheral Nerve Signals
07:30

The Muscle Cuff Regenerative Peripheral Nerve Interface for the Amplification of Intact Peripheral Nerve Signals

Published on: January 13, 2022

2.1K
Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
06:49

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Published on: December 11, 2015

9.0K

Related Experiment Videos

Last Updated: Aug 16, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

4.0K
The Muscle Cuff Regenerative Peripheral Nerve Interface for the Amplification of Intact Peripheral Nerve Signals
07:30

The Muscle Cuff Regenerative Peripheral Nerve Interface for the Amplification of Intact Peripheral Nerve Signals

Published on: January 13, 2022

2.1K
Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
06:49

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Published on: December 11, 2015

9.0K

Area of Science:

  • Robotics
  • Artificial Intelligence
  • Human-Computer Interaction

Background:

  • Wearable exoskeleton robots offer significant potential for assisting human movements in various daily tasks.
  • Real-time activity recognition is crucial for improving the control and assistance capabilities of these robots.
  • Integrating inertial measurement units (IMU) and rotary encoders provides rich data for activity signal analysis.

Purpose of the Study:

  • To implement and evaluate a real-time activity recognition system for exoskeleton robot wearers.
  • To leverage deep learning models for accurate classification of human activities.
  • To optimize and deploy these models on an edge device for practical, real-time application.

Main Methods:

  • Trained and evaluated five deep learning models using activity signals from IMUs and rotary encoders on an exoskeleton robot.
  • Selected and optimized a subset of deep learning models for deployment on an edge device.
  • Tested the system in a continuous action environment recognizing eight common human activities.

Main Results:

  • Achieved an average accuracy of 97.35% for recognizing eight common human activities in real-time tests.
  • Demonstrated a fast inference time of under 10 ms per activity recognition.
  • Reported an overall latency of 0.506 s per recognition on the selected edge device.

Conclusions:

  • The developed real-time activity recognition system effectively supports exoskeleton robot control.
  • Deep learning models deployed on edge devices offer a viable solution for enhancing robot assistance in daily tasks.
  • High accuracy and low latency achieved in real-time tests validate the system's practical applicability.