Bilateral step length estimation using a single inertial measurement unit attached to the pelvis

Area of Science:

• Biomechanics
• Wearable technology
• Gait analysis

Background:

• Accurate estimation of spatio-temporal gait parameters is crucial for physical activity monitoring and clinical applications.
• Previous methods for step length estimation using a single inertial measurement unit (IMU) have limitations due to simplified gait models or linear regression.
• A novel approach is needed to improve the accuracy of gait parameter estimation from IMU data.

Purpose of the Study:

• To propose and validate a new method for estimating bilateral step length and traversed distance during level walking using a single IMU.
• To overcome the limitations of previous methods by directly integrating acceleration data along the progression direction.
• To enhance the accuracy of gait analysis in both clinical and physical activity monitoring contexts.

Main Methods:

• A single IMU was attached to the pelvis.
• The Kalman and Optimally filtered Step length Estimation (KOSE) method was developed, combining a Kalman filter with optimally filtered direct and reverse integration of IMU signals.
• A correction for IMU displacement due to pelvic rotation was implemented, and the method was validated against a stereo-photogrammetric system with nine subjects walking at varying speeds.

Main Results:

• The KOSE method estimated step length with less than 3% error across all subjects.
• Traversed distance was assessed with an error of less than 2%.
• The results demonstrate high accuracy compared to existing single IMU methods.

Conclusions:

• The proposed KOSE method offers superior accuracy for estimating step length and traversed distance from a single IMU compared to existing literature.
• The method's accuracy suggests its suitability for daily monitoring activities in healthy subjects.
• This advancement has significant implications for non-invasive gait analysis and monitoring.