Adaptive Neuro-Fuzzy Inference System for a Three-Wheeled Omnidirectional Mobile Robot

Abstract

This paper presents an Adaptive Neuro-Fuzzy Inference System (ANFIS)-based controller for trajectory tracking of a three-wheeled omnidirectional mobile robot. The strength of the ANFIS-based controller lies in the fact that it integrates the principles of neural networks and fuzzy logic, and hence, it has the potential of capturing the benefits of both in a single framework. The ANFIS-based controller is kinematic-based, and thus a forward kinematic model of the mobile robot is constructed. The effectiveness of the proposed ANFIS-based controller is evaluated in a nonlinear simulation environment. Simulation results showed that the ANFIS-based controller indeed outperformed a well-tuned nonlinear sliding mode controller.