Intelligent backstepping sliding-mode control using recurrent interval type 2 fuzzy neural networks for a ball robot with a four-motor inverse-mouse ball drive

Abstract

This paper presents an intelligent adaptive backstepping sliding-mode control using recurrent interval type 2 fuzzy neural networks (RIT2FNN) for motion control of a ball robot with a four-motor inverse mouse-ball driving mechanism actuated by four independent brushless motors simultaneously. The RIT2FNN is used to on-line learning the uncertain part during the controller synthesis. An adaptive backstepping sliding-mode control together with RIT2FNN is proposed to accomplish robust self-balancing, position control and trajectory tracking of the robot in the presence of mass variations, viscous and Coulomb frictions. Computer simulations are conducted for illustration of the effectiveness of the proposed control method.