Model Predictive Control implementation for MIMO system in Presence of Soft Constraints and Non-linear Disturbance

Abstract

Model Predictive Control (MPC) uses optimization to compute the control signal in presence of system constraints. Therefore, it has gained enormous popularity in various industrial applications because of its constraints management capability. Although, MPC takes large computational time to optimize the control vector and thus it suits for regulating slow-dynamic SISO plants. Here MPC is implemented to regulate fast-dynamic MIMO plant i-e rotors of a quadcopter to exhibit the rotational three degrees of freedom. The performance is evaluated in presence of soft constraints along with linear non-linear sinusoidal external disturbance in terms of percentage overshoot and steady-state error. The simulations are presented by driving the mathematical model of DC motor and after that it is simulated in MATLAB as a MIMO system for four quad rotors. The results are presented in three modes i-e roll, pitch and yaw with four cases each i-e without constraints, with constraints, with constraints and linear disturbance and with constraints and non-linear disturbance. The simulation results have proved that MPC controls the MIMO system comprehensively without breaching the soft constraints.