A new on-line self-adapting fuzzy controller design using unidimensional input-output with dynamical membership functions.

The article focuses on the development of a novel self-adapting fuzzy logic controller (SAFLC) designed to control nonlinear systems using a single one-dimensional input variable—either the previous output or the reference model output—rather than traditional error and error variation inputs. This approach dynamically updates the membership function (MF) boundaries online, enabling the controller to adapt to changing system dynamics without requiring offline optimization or expert-constructed rule bases. The SAFLC was tested on two nonlinear systems—a nonlinear internal system and a sun tracker system—demonstrating superior tracking accuracy, zero overshoot, and robustness to parameter disturbances compared to conventional fuzzy logic controllers. The method reduces computational complexity by minimizing the number of input variables and dynamically tuning MF parameters, offering improved stability and control performance while allowing future enhancements through offline optimization and extension to other fuzzy system types.