Gait parameter adaptation for lower-limb exoskeletons

The field of exoskeletons and assistive orthotic devices is a multidisciplinary issue in the halfway between medicine and robotics. Within the robotic discipline, bipedal robot gaits are generated as a function of parameters such as stride length, foot clearance and body height. These features allow to adapt the gait to different surface characteristics. However, biped robot gaits do not look as natural as human gaits. On the other side, gaits imposed in actual active orthoses are fixed following the Clinical Gait Analysis (CGA) pattern. These CGA patterns ensure a comfortable, natural and safe gait, but do not exhibit flexibility to change some gait features to adapt to the characteristics of the terrain, such as slopes or small obstacles. This paper presents the development of an adaptable gait pattern for a full lower limb active orthosis. Based on an impedance control approach, gait parameters such as step height, body height or step length are modified online, providing a safe and smooth gait pattern. Two shoe insole pressure measurement systems provide ground reaction force and center of pressure to adapt these gait parameters online. To avoid abrupt movements that would affect user comfort, references on position and speed must be smooth, that is, derivable when the characteristics of the gait change. To ensure that position and speed references follow a derivable path, with a very low computational cost, the change from one reference path to another has been modulated with Gaussian windows. Thus, a more adaptable orthosis with soft motion is achieved.