An algorithm to decompose ground reaction forces and moments from a single force platform in walking gait

Area of Science:

• Biomechanics
• Gait Analysis
• Human Movement Science

Background:

• Accurate measurement of individual limb ground reaction forces (GRF) is challenging during the double support phase of walking when using a single forceplate.
• Existing methods struggle to reliably differentiate right and left limb GRF data when summed during this phase.

Purpose of the Study:

• To propose and validate an adaptive transition function for decomposing and estimating individual right and left GRF and moments during the double support phase of walking.
• To develop a method that accounts for walking speed and GRF characteristics.

Main Methods:

• An adaptive transition function incorporating a shape coefficient, optimized using single support GRF parameters via non-linear least-square curve-fitting.
• Multiple regression analysis to identify key GRF parameters for decomposition.
• Validation using relative Root Mean Square Error (RMSE) and differences in Center of Pressure (CoP).

Main Results:

• Low relative RMSE values across different force and moment components during double support (e.g., 3.8% for vertical force, 4.3% for frontal moment).
• Maximum GRF differences normalized to body mass were below 1 N/kg.
• Mean CoP difference between real and decomposed signals was 0.0135 m.

Conclusions:

• The proposed adaptive transition function accurately decomposes right and left GRF and moments during the double support phase.
• This method provides a valuable tool for gait analysis in single forceplate configurations, enhancing the discernment of individual limb dynamics.