Neural network based fractional order sliding mode tracking control of nonholonomic mobile robots

Abstract

In this study, the position tracking control problem of a nonholonomic mobile robots with system uncertainties and external disturbances is examined. In the design approach, a fractional-order sliding surface is presented that offers asymptotic stability of the system states towards their equilibrium points. A fractional order sliding mode controller is developed based on the presented sliding surface in order to handle system uncertainties and external disturbances in a robust manner. A radial basis function neural network is used to approximate the nonlinearities of the dynamic structure. The weighted matrices of neural networks are updated in an online mode. The controller’s adaptive bound portion is used to manage neural network reconstruction error and provide upper bounds on disturbances and uncertainty. Using the Lyapunov technique and Barbalat’s Lemma, the asymptotic stability of the control system is evaluated. Moreover, a numerical simulation study is carried out to illustrate the effectiveness of the proposed control approach by comparing the results with the existing control approaches.