Fig. 4

Effects of aging on responses to exoskeleton assistance. (A) Measured optimized exoskeleton torque scaled to body mass. (B) Exoskeleton ankle angle during assisted walking. Positive angle corresponds with plantarflexion. (C) Exoskeleton ankle power, scaled to body mass. (A-C) Black lines and gray shaded regions show average values and ± 1 standard deviation for older adults. Blue and green lines show average values for younger adults in human-in-the-loop optimization studies targeting faster walking speed [11] and lower metabolic rate [16], respectively. (D) Changes in normalized step frequency variability as participants adapt to exoskeleton assistance. Exponential (exp.) model fits are plotted in solid lines for the study duration and extended to asymptotes by dashed lines. 95% confidence intervals are shaded in gray and green for older and younger adults, respectively. Younger adult trends are recreated from [32]. (E) Cost of transport changes between unassisted (zero torque) and assisted walking, observed across participant ages. (F) Age-related changes in walking speed. Average self-selected overground speed (black line), and 95% confidence interval (gray shaded region) are recreated from [44]. Older adult participants are plotted in unique colors