Path following of ship based on sliding mode control with improved RBF neural network and virtual circle

To address the unmeasured velocity, external disturbance and internal model uncertainty for following path of an under-actuated ship, paper presents a sliding mode control method based on radial basis function(RBF) neural network velocity observer. enhance RBF performance approximating unknown, arc tangent function was exploited in to update its weight values. Then, nonlinear observer built via hyperbolic deal with ship. Furthermore, order avoid overshoots when ship is moving way points, virtual paths variable circle turning angle were designed at joints capability. Finally, simulation results show that controller can force follow accurately reference case time-varying disturbances without measured accuracy network, thus demonstrating effectiveness.