The design of a robust maneuver load alleviation (MLA) system for a high-performance aircraft is studied in this paper. First, the aeroservoelastic (ASE) models of a high-performance military aircraft in climbing maneuver at varying Mach numbers are established. Then, a linear parameter-varying (LPV) model of the ASE systems is constructed and an [Formula: see text] robust controller is designed based on the LPV model. The robust control is realized via a pair of outboard ailerons to alleviate the wing-root bending moments in the climbing maneuvers. To compensate the loss of performance in the load alleviation, a controller based on recurrent neural networks is designed in the flight control. Finally, some numerical simulations are made to testify the performance and robustness of the MLA system.