In this paper, a novel auxiliary saturation compensation system is constructed to eliminate the negative effect of asymmetric nonlinear actuator saturation (ANAS). Two chatter-free neuroadaptive sliding mode tracking control strategies are proposed by combining nonsingular terminal sliding mode (NTSM) control technique with radial basis function neural network (RBFNN) for high speed trains for resp., the actuator saturation-free and ANAS saturation cases. These two control strategies are used to guarantee that the closed-loop signals are bounded and the steady-state tracking errors converge to a residual around zero. The common assumption that the initial velocities of the train and the desired trajectory are zero, is removed in this paper. In addition, the computationally inexpensive strategy contains an optimized NN adaptation mechanism where only one parameter requires online updating no matter how many neurons are chosen. Finally, numerical examples are provided to demonstrate the validity of the proposed control strategies.
Read full abstract