Error-state Kalman filter for lower-limb kinematic estimation: Evaluation on a 3-body model.

Area of Science:

• Biomechanics
• Human Movement Analysis
• Wearable Sensor Technology

Background:

• Accurate human lower-limb kinematic measurements are vital for gait analysis, injury prevention, and performance enhancement.
• Existing methods often face challenges with accuracy and confounding factors like soft tissue artifacts.

Purpose of the Study:

• To present a new method for estimating lower-limb kinematics using an error-state Kalman filter (ErKF) with body-worn inertial measurement units (IMUs) and kinematic constraints.
• To evaluate the ErKF method's accuracy on a simplified 3-body lower-limb model in both simulated and experimental settings.

Main Methods:

• Utilized an error-state Kalman filter (ErKF) integrating data from an array of body-worn inertial measurement units (IMUs).
• Incorporated four kinematic constraints into the filtering process.
• Validated the method on a 3-body lower-limb model (pelvis and two legs) during walking, comparing against simulation and optical motion capture (MOCAP).

Main Results:

• Root Mean Square (RMS) differences for hip joint angles were below 0.2 degrees (simulation) and 1.4 degrees (MOCAP).
• RMS differences for stride length and step width were within 1% and 4% (simulation), and 7% and 5% (MOCAP), respectively.
• The simplified model evaluation minimized confounding factors, allowing direct assessment of the ErKF method.

Conclusions:

• The developed ErKF method provides highly accurate lower-limb kinematic estimations.
• Results indicate the method's potential for future application in more complex human movement models.
• This approach shows promise for advancing biomechanical analysis in diverse fields, from sports to healthcare.