Robust model-free control of a class of uncertain nonlinear systems using BELBIC: stability analysis and experimental validation

The applicability of a brain emotional learning-based intelligent controller (BELBIC) is based on the emotion processing mechanism in brain, sensory inputs and emotional cues. The inspiration of this intelligent controller is the computational model of limbic system in the mammalian brain. In the thematic literature, some stability analyses for BELBIC are presented. Most of them are dealing with some specific systems or mathematical models of these systems are required. In this study, a more general stability analysis for the BELBIC is proposed that can be simply applied to the tracking control of a large group of nonlinear systems. The novelty and main contributions of this study are treating the BELBIC as an uncertainty estimator to pave the way for stability analysis. The control law comprises a model-free control approach. The satisfactory performance of the proposed method is proved by numerical simulations on an induction motor and comparison with a neuro-fuzzy controller. The considerable influence of the BELBIC in uncertainty estimation and reducing the tracking error is also verified by an experimental study on a SCARA manipulator. In addition, to highlight the superiority of BELBIC in practical implementations, a comparison with a controller using an adaptive uncertainty estimator has been presented.