Decomposition of superimposed ground reaction forces into left and right force profiles.

The process of collecting ground reaction force data by mounting a forceplate beneath a treadmill belt has the advantage that numerous walking trials can be analyzed without the problem of subjects 'targeting' their footsteps. However, a potential problem is that the measured forces represent a summation of bilateral force profiles during the double support phase of walking. To address this issue, an algorithm is described for decomposing superimposed ground reaction force data into individual left and right profiles. It is based on an examination of the side-to-side oscillations of the measured center of pressure (CoP). Whenever the measured CoP exceeds a certain threshold, it is assumed that the person is being supported by a single limb, and the measured GRF data reflect the forces under that limb. Conversely, when the measured CoP indicates that both feet are on the treadmill, it is assumed that the location of the individual CoP under each foot is given by wL2 and wR2. These quantities reflect the greatest excursion of the measured CoP towards the left and right sides of the forceplate, respectively. With this assumption, individual GRF profiles can be calculated by means of solving two simultaneous equations--one describing the equilibrium of forces in the vertical direction, and one describing the equilibrium of moments about an antero-posterior axis of the forceplate. The algorithm describing this procedure is simple enough to be implemented on a spreadsheet and yields estimates for average force, impulse, peak force and stance time that are typically within 3% of the true values.

[1]  Peter R. Cavanagh,et al.  Design principles for a zero-gravity locomotion simulator , 1989 .

[2]  R. Kram,et al.  A treadmill-mounted force platform. , 1989, Journal of applied physiology.

[3]  E. Chao,et al.  Biomechanical analysis of foot function during gait and clinical applications. , 1983, Clinical orthopaedics and related research.