Model Predictive Sliding Control for Finite-Time Three-Axis Spacecraft Attitude Tracking

This paper deals with the robust optimal three-axis attitude tracking problem of spacecrafts. To this aim, the exceptional features of two advanced control approaches, i.e., nonlinear terminal sliding mode control (TSMC) and linear model predictive control (MPC), are combined in a compound double layer structure. The proposed approach addresses the global stability and robust attitude tracking of near-polar orbit spacecrafts actuated by a set of reaction wheels and subject to unknown disturbances, uncertainties, and actuators' saturation. The TSMC controller is designed to guarantee the robust three-axis attitude tracking goal in the presence of disturbances and uncertainties provided that proper target points are available. The next part of the compound controller takes the advantages of the MPC technique (i.e., preknown attitude maneuvers and constraints) into account to provide the required optimal target points for the lower level TSMC controller. Despite the fact that the spacecraft has nonlinear nature, we have analytically derived conditions under which the actuators' constraints satisfaction is guaranteed a priori. Finally, a set of simulation results are provided to illustrate the effectiveness and performance of the proposed method.