In this paper, we present a novel interval-valued fuzzy model-based controller for handling the effects of uncertainty in controlling a complex dynamical system. Theoretically, model-based controllers may be the ideal control mechanisms; however, they are highly sensitive to model uncertainties and lack robustness. These controllers are also computationally intensive, rendering them unusable for many real-world applications. In this work, we incorporate an interval fuzzy logic paradigm into a computed-torque controller for a 3-PSP parallel robot. This paradigm aims to handle the uncertainties in the robot model. The proposed approach benefits from algebraic operations on type-I fuzzy numbers to enhance its capability in dealing with uncertainty. The simulations prove the superiority of the proposed controller in the presence of uncertainty. Furthermore, comparisons with a competing type-I reduced controller as well as a PD controller show this superiority to be more pronounced especially when noise level is remarkably high. Moreover, the designed controller satisfies the computational complexity constraints for real-time implementation.

Keywords