B-spline-based neuro-fuzzy velocity field navigation and control for a nonholonomic mobile robot

A neuro-fuzzy tuned velocity potential field navigation controller for nonholonomic mobile robot in dynamic and partially known environments containing moving obstacles is presented. The navigation consider the nonholonomic constraints and is based on building velocity potential field using robot and obstacle´ velocity information. In order to handling unknown factors in the navigation, such as disturbance and nonlinear dynamics, a B-spline fuzzy networks (BSFN) approximation-based computed-torque controller is proposed. The BSFN weights are tuned on-line. The stability of the overall closed-loop system, composed by the nonlinear robot dynamics and the robust neuro-fuzzy controller, is guaranteed by the Lyapunov theory.