The uncertain optimization problem for structural dynamic responses of artillery systems is studied from the perspective of uncertainty and system engineering. First, a rigid–flexible coupling dynamic model of artillery at the maximum firing angle is constructed, which can achieve coupling relationships between interior ballistics, launching loads and artillery movement. Secondly, an integration method of optimal Latin hypercube design and regression analysis is used to obtain the polynomial model, followed by sensitivity analysis. Thus, the key parameters affecting the dynamic response are identified. An interval uncertain optimization method for artillery structural dynamic responses considering robustness and interval economy is then proposed, based on nonlinear interval programming and the nested optimization solving strategy, which integrates the back-propagation neural network with the genetic algorithm as the inner optimizer, and uses the non-dominated sorting genetic algorithm-II as the outer optimizer. Finally, an example is presented to demonstrate the validity of the proposed method.