Neural Network-Based Sliding Mode Control for Dual-Motor Servo Systems

This paper proposes a fast terminal sliding mode (FTSM) control for dual-motor servo systems with external output disturbance. The method of neural network (NN) is used for FTSM, such that the system is sufficient to achieve the elimination of external output disturbance and tracking control for the reference signal with prescribed performance. A continuous differentiable function is adopted to account for the non-differential deadzone. The proposed FTSM has a faster convergence rate than the conventional sliding mode. More importantly, the tracking error is uniformly ultimately bounded. The simulation results are included to verify the reliability and effectiveness.