Natural gait event-based level walking assistance with a robotic hip exoskeleton

In this paper, we present an assistance strategy for level walking by using a robotic hip exoskeleton. Our strategy utilizes a foot contact event estimated by an inertial measurement unit (IMU) on the pelvis. The gait cycle is composed of three phases. The transitions between the phases are established upon natural gait events that are inevitable and perceived reliably by sensors attached to our exoskeleton. The presented strategy provides a systematic way of adjusting the quantity of assistance that corresponds to the wearer's preference and needs, and also provides explicit principles for the initiation and termination of assistance. When a step begins, the maximum torque and duration of assistance are decided, and the torque profile for the entire step is designed in advance. We conduct experiments in order to investigate the effect on metabolic cost when walking on a motorized treadmill.