Kalman Filter-Based MPC Control Design and Performance Assessment of MIMO System

In this paper, a Kalman filter parameter updated model established controller is scheduled for multiple input multiple output (MIMO) system through constraints. Since model/plant mismatches occur in real-world plants due to uncertainties, so the constrained tracking is very important. The plant model state parameters are estimated using Kalman filter, and the parameters are updated in the controller. To delineate the attainment of the scheduled Kalman filter-based model predictive control (MPC) approach is compared with phase and gain margin and biggest log modulus tuning of P and I value of the PI controller. The proportional and integral controller settings of the plant/system are found two ways using multi-loop method. Firstly, the system controller settings tuned are established on Biggest Log Modulus tuning method, which is also called detuning method. Secondly, the controller settings are tuned from the gain and phase margin requirements of plant/system by using the derived formulae in the literatures. The performance and transient response characteristics of the control schemes are compared to exhibit the high performance of the scheduled method for the bench mark MIMO systems using MATLAB simulation software.