Automatic tuning with feedforward compensation of the HuREx rehabilitation system

Abstract

The Human inspired Robot Exoskeleton (HuREx) is a device designed for lower limb rehabilitation for people with stroke or disabilities. In order to provide soft human-robot interaction and improve tracking performance, an impedance controller with model based feedforward (FF) compensation was implemented. The trajectory tracking performance was compared with and without the FF compensation. The results show that it was not possible to completely eliminate the tracking error from the standard impedance controller and it was prone to noise. On the other hand, the FF compensated system was able to track the desired position almost perfectly, proving the advantage of the use of model-based control strategy. Iterative Feedback Tuning (IFT) was employed to automatically tune the inner force control loops (of the pneumatic muscle actuators) for both control strategies. The algorithm has worked very effectively and has successfully optimised the controller gains, hence proving its feasibility to adapt to changes in system parameters and different wearers.