Direct force control of upper-limb exoskeleton based on fuzzy adaptive algorithm

Abstract

In order to synchronize human and machine positions and minimize human-machine interaction forces in exoskeleton control, we present a two-degree-of-freedom (2-DOF) upper-limb exoskeleton model with power enhancement and direct force control strategy based on fuzzy adaptive algorithm. The conventional PD controller is widely used in exoskeleton control because it is model independent and its gains can be easily tuned. However, the speed of movement of the operator and the mass of external load are uncertain in practice; hence, the parameters of a conventional PD controller have to be adjusted according to the velocity of the motion and external loads to ensure the effectiveness of trajectory tracking. Additionally, there is a lag in the response time when the operator starts to move or changes direction suddenly. Therefore, this study proposes the use of an adaptive controller combining the fuzzy set techniques and PD controller to improve trajectory tracking. Robustness testing of the fuzzy PD controller for the external load uncertainty and motion velocity change are also investigated. The simulation results clearly indicate the superior performance of the fuzzy adaptive PD controller over the conventional one for tracking performance with external load uncertainty and motion velocity variance.