Adaptive Neural Network Back-Stepping Control for Servo Systems with Current Constraints

In this paper, an adaptive neural network control (ANNC) scheme is proposed for the servo system with unknown uncertainties and current constraints. First, a novel system transformation scheme is designed to transform the servo system with current constraints into the equivalent “unconstrained” one. Second, to restrict the activation state of the constraint mechanism, a security boundary is incorporated into the designed strategy. Moreover, a nonlinear control scheme based on the ANNC and the extended state observer (ESO) is created to compensate for the system disturbances utilizing the modified model and back-stepping technology. The stability analysis of the closed-loop system is rigorously given by the Lyapunov stability theory. Finally, the effectiveness and feasibility of the proposed control scheme have been validated through simulation studies on the motor drive system.