Drive-augmented model predictive control with Kalman state observer for long-stroke maglev planar motor

This paper proposes a model predictive control method for magnetic levitation planar motors with hysteresis compensation of the drive system to achieve high responsiveness and high accuracy long stroke positioning of planar motors. Based on the dynamics model of the planar motor developed in the laboratory, and the explicit model of the drive system, including the driver and the coil, an augmented generalized model is built as a predictive model to predict the future position of the motor by combining the actual measured position, state estimation, and perturbation estimation via a Kalman state observer, and an explicit analytical solution for optimal control is obtained by minimizing the cost function. Finally, the effectiveness of the proposed model predictive control method in realizing high-accuracy time-varying positioning of magnetic levitation planar motors is experimentally verified.