A multi-model robust guaranteed cost soft switching control method for a near space vehicle with variable wing
sweep angle is proposed. The work space of the nonlinear system is divided into multiple regions. Then local T-S fuzzy models are built and local robust guaranteed cost controllers are designed according to the performance index by using descriptor system method in each region. Consequently, the calculated amount is greatly reduced. In addition, the system states are smooth in the switching process due to the blurred boundaries. Finally, the proposed method is adopted to control the attitudes of near space vehicle during its wing morphing, and simulation results show the effectiveness of the proposed method.