Movement control based on model predictive control using Kalman filter for known and unknown noise covariance matrices

This paper proposes two motion control algorithms, one based on model predictive control (MPC) and traditional Kalman filter control algorithm, and the other based on MPC and adaptive Kalman filter control algorithm. Both control algorithms consider the influence of noise and are respectively used to solve the problem where the noise covariance matrix is completely known or completely unknown. Under the influence of noise, it is difficult for general MPC to achieve ideal control effects. In contrast, the proposed MPC algorithms filtered by traditional Kalman filters and adaptive Kalman filters have strong robustness and anti-interference ability. Finally, the control algorithms proposed in this paper are simulated in the height control of unmanned aerial vehicles through mathematical modeling, and the feasibility of the control algorithms in zero steady state and non-zero steady state is verified.