An adaptive and robust online method to predict gait events.
Summary
A new Gait Phase Estimator (GPE) accurately predicts walking timing and stride duration, adapting to speed changes and irregular strides for better assistive device function.
Area of Science:
- Biomechanics
- Robotics
- Human-Computer Interaction
Background:
- Accurate gait phase timing is crucial for assistive devices to optimize walking efficiency and reduce metabolic cost.
- Current timing methods struggle with variations in stride duration due to speed changes and cannot effectively handle abnormal strides.
Purpose of the Study:
- To introduce a novel Gait Phase Estimator (GPE) for predicting temporal gait events and stride duration.
- To evaluate the GPE's prediction accuracy, robustness to disturbances, and adaptability to speed variations.
Main Methods:
- The Gait Phase Estimator (GPE) utilizes a weighted forward moving-average of stride duration for predictions.
- Experimental evaluation involved three subjects walking on a treadmill at three different speeds.
Main Results:
- The GPE demonstrated superior prediction performance compared to predefined estimates on average.
- The GPE automatically adapted to walking speed changes and showed robustness to irregular strides.
Conclusions:
- The proposed GPE offers improved and simplified stride duration estimation.
- This method has significant potential to enhance the performance of gait-assistive devices and experimental protocols.
