Control strategies for human-inspired robotic exoskeleton (HuREx) gait trainer

Abstract

A Human-inspired Robotic Exoskeleton (HuREx) gait trainer which utilizes an antagonistic pair of pneumatic muscle actuators (PMAs) has been developed. An accurate and robust control system with high bandwidth is needed to enable clinical testing of the HuREx on patients. This paper presents two papers implemented on the redeveloped robotic platform. One is an impedance based controller. The other is a robust sliding mode controller. Trajectory tracking tests with SMC showed faster tracking performance with good accuracy compared to the impedance based controller. Robustness of the SMC is also investigated with deliberately introduced modelling inaccuracy, through perturbations in the model parameters. This is used to show that the system can react when there are uncertainties in the real system, such as when a real patient is wearing the device. The contribution of this research is to demonstrate that a robust SMC can accurately and robustly control the HuREx device in the range required for gait rehabilitation.