An aeroelastic optimization design methodology for air vehicle considering the uncertainties in maneuver load conditions is presented and applied to a structural design process of low-aspect-ratio wing. An aerodynamic load correction model is developed and used to predict the critical load conditions with the perturbations of theoretical linear aerodynamic forces and experimental aerodynamic forces from wind-tunnel test, when concerning the uncertainties in use of theoretical linear and experimental aerodynamic forces. Three objective functions of critical loads are defined. The load evaluations for three wing sections are investigated in four characteristic maneuvers, and the most critical load conditions are confirmed by using the sequential quadratic programming method. On this basis, the aeroelastic optimization design employing the genetic-gradient hybrid algorithm is conducted, in which the objective is to minimize structural mass subject to the constraints of stress, deformation and flutter speed. The resulting optimal structure is heavier than the one simply based on the theoretical linear or experimental aerodynamic forces. However, it is more robust when encountering the critical load conditions in actual flight due to the consideration of uncertainties in aerodynamic forces in the early design phase, thereby, the risk of structural redesign can be reduced. An aerodynamic load correction model is developed and used to predict the critical load conditions with the perturbations of theoretical linear aerodynamic forces and experimental aerodynamic forces from wind-tunnel test, when concerned the uncertainties in use of theoretical linear and experimental aerodynamic forces. Three objective functions of critical loads are defined. The load evaluations for three wing sections are investigated in four characteristic maneuvers, and the most critical load conditions are confirmed by using the sequential quadratic programming method. On this basis, the aeroelastic optimization design employing the genetic-gradient hybrid algorithm is conducted, in which the objective is to minimize structural mass subject to the constraints of stress, deformation and the flutter speed. The resulting optimal structure is heavier than the one simply based on the theoretical linear or experimental aerodynamic forces. The resulting optimal structure is heavier than the one only based on the theoretical linear or experimental aerodynamic forces. early design phase, thereby, the risk of structural redesign can be reduced.