An Indirect Type-2 Fuzzy Neural Network Optimized by the Grasshopper Algorithm for Vehicle ABS Controller

Abstract

Model nonlinearity, structured and unstructured uncertainties as well as external disturbances are some of the most important challenges in controlling the wheel slip in moving vehicles. Based on the interval type-2 fuzzy neural network, we construct an indirect exponential sliding-mode (ESM) controller for improving the performance of vehicle antilock braking systems (VABSs) in the face of the uncertainties. Lyapunov stability postulate is used to verify the stability of the closed-loop system and also to extract the adaptation rules. In this scheme, the reaching law for the sliding surface is regulated based on an exponential surface in order to eliminate the produced chattering. Selecting appropriate controller constants and adaptation rules leads to quicker signal convergence and a better management of the control signal restrictions. These constants are optimized by defining a cost function and employing the grasshopper optimization algorithm (GOA) to search for an optimal solution. Thus, we provide an optimized robust adaptive indirect ESM controller with GOA for VABS. The efficacy of the proposed method is verified by analyzing the obtained results and comparing its performance with some other control schemes for various road conditions and driving maneuvers. The results of this work affirm that the designed method makes a significant improvement in the performance of VABS control.