An Adaptive Wearable Parallel Robot for the Treatment of Ankle Injuries

This paper presents the development of a novel adaptive wearable ankle robot for the treatments of ankle sprain through physical rehabilitation. The ankle robot has a bioinspired design, devised after a careful study of the improvement opportunities in the existing ankle robots. Robot design is adaptable to subjects of varying physiological abilities and age groups. Ankle robot employs lightweight but powerful pneumatic muscle actuators (PMA) which mimics skeletal muscles in actuation. To address nonlinear characteristics of PMA, a fuzzy-based disturbance observer (FBDO) has been developed. Another instance of an adaptive fuzzy logic controller based on Mamdani inference has been developed and appended with the FBDO to compensate for the transient nature of the PMA. With the proposed control scheme, it is possible to simultaneously control four parallel actuators of the ankle robot and achieve three rotational degrees of freedom. To evaluate the robot design, the disturbance observer, and the adaptive fuzzy logic controller, experiments were performed. The ankle robot was used by a neurologically intact subject. The robot–human interaction was kept as active–passive while the robot was operated on predefined trajectories commonly adopted by the therapists. Trajectory tracking results are reported in the presence of an unpredicted human user intervention, use of compliant and nonlinear actuators, and parallel kinematic structure of the ankle robot.