Model predictive control of a dual-stage actuator system for fast setpoint tracking

Dual-stage actuator (DSA) servomechanisms have attracted considerable industrial applications for the benefits of both large working range, high positioning accuracy and fast response, which are typically difficult to achieve simultaneously by a single-stage actuator. However, the control design of such systems suffers from hard constraints due to actuator input limits and position output profile requirements. For setpoint tracking tasks, the controllers with the merit of easy implementation have been reported but come at the cost of performance limitations such as the risk of system instability due to ignoring constraints during the design process or the requirement of retuning the control parameters for different setpoints. This paper presents a model predictive control (MPC) method to obtain an universal DSA controller, which explictly considers the design constraints during the design optimization and does not require retuning for different setpoints. Simulations are studied to verify that the MPC can offer versatile trajectories for each actuator and achieve more robust tracking performance with respect to different setpoints in comparison with the conventional linear feedback-feedforward controller and the existing composite nonlinear feedback controller.