Design of Model Predictive Controller Based on Extended State Observer

This paper proposes a linear model predictive control (LMPC) based on extend state observer (ESO) which can approximate the control performance of the nonlinear model predictive control (NMPC). The ESO is used to estimate the disturbance caused by the linearization of actual nonlinear system. The estimated disturbance is added into the system model of the LMPC controller design. The low speed control problem of the permanent magnet DC (PMDC) motor is studied. The simulation results show that this method can obtain satisfied control performance and the ESO can estimate the uncertain disturbance accurately caused by the linearization if the parameters of ESO are chosen appropriately.