Robust Model Predictive Control Based on MRAS for Satellite Attitude Control System

Abstrak

In this paper, an improved robust model predictive controller (RMPC) is proposed based on model reference adaptive system (MRAS). In this algorithm, using the MRAS a combinational RMPC controller for three degree freedom satellite is designed such that the effect of moment of inertia uncertainty and external disturbance is compensated on the stability and performance of closed loop system. Control law is a state feedback which its gain is obtained by solving a convex optimization problem subject to several linear matrix inequalities (LMIs). To avoid the actuators saturation an input constraint is incorporated as LMI in the mentioned optimization problem. In addition to, using the MRAS system the effect of input disturbance is rejected on the system.The advantages of this algorithm are needless to exact information from system’s model, robustness against model uncertainties and external disturbance. Results from the simulation of the system with the proposed algorithm are presented and compared to generalized incremental model predictive control (GIPC). The results show that the suggestive controller is more robust than the GIPC method.DOI:http://dx.doi.org/10.11591/ijece.v4i1.4961