Related Experiment Video
Updated: Oct 17, 2025
11:16
Engineering Platform and Experimental Protocol for Design and Evaluation of a Neurally-controlled Powered Transfemoral Prosthesis
Published on: July 22, 2014
16.4K
Real-Time Gait Phase Estimation for Robotic Hip Exoskeleton Control During Multimodal Locomotion
Inseung Kang1, Dean D Molinaro1,2, Srijan Duggal1
1School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, GA, 30332 USA.
IEEE Robotics and Automation Letters
|October 7, 2021
Summary
This study introduces a novel gait phase estimator using a convolutional neural network for robotic hip exoskeletons. The new method enhances real-time control during diverse walking conditions, improving exoskeleton assistance.
Area of Science:
- Robotics
- Biomechanics
- Machine Learning
Background:
- Gait phase estimation is crucial for controlling robotic exoskeletons during cyclic tasks.
- Conventional methods struggle with dynamic changes in walking speed and locomotion modes.
- Accurate gait phase estimation is needed for seamless exoskeleton assistance in real-world scenarios.
Purpose of the Study:
- To develop and validate a robust gait phase estimator for real-time robotic hip exoskeleton control.
- To improve exoskeleton assistance adaptability during multimodal locomotion.
- To overcome limitations of traditional gait phase estimation methods in overground walking.
Main Methods:
- Development of a convolutional neural network (CNN)-based gait phase estimator.
- Validation of the estimator's performance during multimodal locomotion.
- Comparison with conventional time-based gait phase estimation techniques.
Main Results:
- The CNN-based estimator accurately predicted gait phase during multimodal locomotion without mode-specific information.
- Achieved a root mean square error (RMSE) of 5.04 ± 0.79% for gait phase estimation.
- Significantly outperformed the literature standard (p < 0.05), demonstrating superior accuracy.
Conclusions:
- The developed CNN gait phase estimator shows significant promise for real-time control of robotic hip exoskeletons.
- This technology enables more natural and seamless navigation through varied terrain settings.
- Highlights the potential for advanced exoskeleton applications in realistic, dynamic environments.
