Force control of an uncertain series elastic actuator system via a fuzzy sliding mode controller and a nonlinear state estimator

Abstract

Nowadays, series elastic actuators play an important role in actuation technology, especially for mechatronics and robotic applications. In this article, the accurate tracking of output force for a series elastic actuator is investigated. In order to achieve this aim, a novel robust control scheme is developed to overcome friction effects, backlash in the drivetrain, measurement noises (sensor inaccuracy) and uncertainties in the dynamic model. The control scheme consists of a fuzzy sliding mode controller and a nonlinear state estimator. The theory of Lyapunov stability is used for the formulation of the proposed controller and also the fuzzy logic strategy is used for tuning the sliding surface parameter in order to achieve accurate and chattering-free performance. Furthermore, a powerful estimator algorithm, unscented H-infinity filter, is added to the control scheme to improve the robustness of the system. Simulation results, as well as experimental analysis, approved the efficiency of the proposed approach in the presence of uncertainties and disturbances.