NARMA-L2-based control of wheeled mobile robots singularly perturbed by using neural networks: Curvilinear approach

Abstract

The present paper uses the learning ability of neural networks (NN) for nonlinear systems in order to design a controller for trajectory tracking problems in wheeled mobile robots that have their kinematic constraints violated. The singular perturbation approach is used to highlight the presence of uncertainties related to the violation of the kinematic constraints and propose an alternative global feedback law in order to regard un-modeled dynamic terms. To this end, a robust control scheme based on NARMA-L2 approach and a modified version of an already existing dynamic backpropagation algorithm are used to improve the tracking error performance. The robustness of the proposed controller is compared with the classic static-state linearization techniques and a converging tracking error toward a small neighborhood containing the origin will be viewed.