| Summary: | In this article, a hybrid control approach combining sliding mode and H-infinity is proposed for an uncertain single-link flexible manipulator. The sliding mode controller stabilizes the nonlinear manipulator system, while the H-infinity controller enhances the noise rejection capability of the system by reducing the total system nonlinearity. The proposed hybrid controller is designed with the goal of rejecting external noises, hence providing a higher system performance, compared to a pure sliding mode controller. To avoid unintentional consequences of switching between the sliding mode and H-infinity controller, a fuzzy neural network weighting method is designed providing a smooth synthesis of both controller outputs. The neuro-fuzzy method applies a weighted combination of the two controller outputs to the manipulator system. In addition, a novel fuzzy estimation method is used to characterize the unstructured nonlinear disturbances in manipulator systems. The proposed hybrid control approach along with the fuzzy estimator is capable of providing a versatile means to stabilize flexible manipulator systems while maintaining a precise reference trajectory tracking in presence of unstructured uncertainty and nonlinear dynamics, as demonstrated by simulation results.
|
|---|