Fig 1.
Influence of modeling choices on predicted gait pattern.
Predicted knee and ankle kinematics (A) and kinetics (B), muscle activations (C), and ground reaction forces (D) with the different models. Experimental data (shaded areas) are shown as mean ± 2 standard deviations. The experimental electromyography data were normalized to peak activations of the new model with high contact spheres and with toes (dashdot orange curve). The vertical black lines indicate experimental stance to swing transition. The old and new models have the same mass but different mass distribution: the new model having a lighter torso, but heavier legs as compared to the old model. The difference in the vertical position of the contact spheres (about 1 cm) is illustrated in (E), and an example toe extension is depicted in (F). Results for all joints and muscles are shown in S1–S3 Figs.
Table 1.
Influence of modeling choices on the metabolic cost of transport (COT).
Fig 2.
Snapshots of the predicted half gait cycles.
Predicted half gait cycle with the old model with low contact spheres and without toes (A; solid black curve in Fig 1) and with the new model with high contact spheres and with toes (B; dashdot orange curve in Fig 1). We can appreciate the difference in knee flexion during early stance. The ground reaction force vector passes close to the knee joint center during early stance for the model without toes (C), whereas it is posterior to the knee joint center for the model with toes (D), requiring knee extension torques.
Fig 3.
Influence of Achilles tendon stiffness on predicted gait pattern.
Predicted knee and ankle kinematics (A) and kinetics (B), triceps surae activations (C), and normalized muscle fiber lengths (D) with models with different Achilles tendon stiffness. Predicted metabolic cost of transport (E) and stride length (F) as a function of the Achilles tendon stiffness (in percent of the nominal value). Experimental data (shaded areas) are shown as mean ± 2 standard deviations. The experimental electromyography data were normalized to peak nominal activations (100% of the generic Achilles tendon stiffness). The vertical black lines indicate experimental stance to swing transition.