This article presents an adaptive filter backstepping control strategy for reusable launch vehicles attitude tracking during reentry phase in the presence of input constraints, model uncertainties and external disturbances. The control-oriented model with uncertainties is constructed, where the uncertainties do not satisfy the linear parameterization assumption. To cope with input constraints, an auxiliary system is introduced, and the states of which are applied to the procedure of control design and stability analysis. Second-order filters are employed to overcome the ‘explosion of terms’ problem inherent in traditional backstepping control. Moreover, the stability of the closed-loop system is proven via Lyapunov technique, and the tracking error can be forced into an arbitrarily small neighborhood around zero (i.e. semi-globally uniformly ultimate bounded tracking). Finally, the 6-degree-of-freedom nonlinear reusable launch vehicle simulation results are presented to verify the effectiveness of the control strategy.