A Bio-Inspired Robotic Orthosis for Gait Rehabilitation

This research presents the modeling and control of a robotic gait training orthosis. The robotic orthosis has six degrees of freedom. Hip and knee sagittal plane rotation are powered by the antagonistic actuation of pneumatic muscle actuators (PMA). The modeling of the PMA was performed using artificial neural network (AAN). A fuzzy logic control scheme was used as a trajectory tracking controller. The purpose of the fuzzy logic control scheme was to guide the human subject's limbs on reference physiological trajectories with minimum trajectory tracking errors. The fuzzy logic control scheme was evaluated on a healthy subject during robot assisted walking and satisfactory performance was obtained.