Fig. 1
From: Optimizing exoskeleton assistance to improve walking speed and energy economy for older adults
Multi-objective human-in-the-loop optimization. Real-time speed and metabolic (met.) data are collected as a participant walks on a self-paced treadmill wearing bilateral tethered ankle exoskeletons (exos). These data are used to evaluate exoskeleton torque parameters according to a multi-objective cost function. An optimizer generates new sets of exoskeleton parameters to be tested, in search of lower-cost parameters. Each parameter set – peak magnitude (Mp), peak time (tp), rise time (tr), and fall time (tf) – defines an exoskeleton torque profile that is applied once per step during walking