<p>This paper proposes a new filtered P-type iterative learning control scheme that is scaled to Hilfer-type fractional stochastic pantograph inclusions, which are a complex group of models with a fractional-order and proportional time delays and stochastic disturbances. The proposed method uses the rich memory characteristics of fractional derivatives by incorporating the concept of fractional calculus within the iterative learning control system so that refinement of the error feedback can be obtained with a series of iterations. Exponential filtering of the tracking error is effective in reducing the high-frequency stochastic noise and this increases the robustness and speed of convergence of the learning algorithm. The comprehensive theoretical study proves that such strategy ensures faster convergence of mean square error and much better resistance to stochastic perturbation and the effect of time delays. Numerical simulations of a stochastic longitudinal trajectory-tracking aerospace model have shown that the proposed control scheme has better tracking performance and robustness even in the presence of significant uncertainties. The application of these findings opens the way to the advanced uses of the concept of fractional stochastic iterative control in aerospace guidance and control systems.</p>