An adaptive backstepping terminal sliding mode controller is designed for ASV with highly nonlinear and uncertain parameters in the reentry phase. Firstly, a strict-feedback nonlinear system with time-varying parameters for the ASV is established. Then, an adaptive strategy is used to estimate the uncertain parameters. Computation explosion is reduced by introducing first-order low-pass filters. In the nth step, the terminal sliding mode control is introduced, which can improve the robustness of system and error convergence rate. The introduced generalized inverse matrix can effectively avoid the singularity in the parameter estimation of control gain matrix process. All errors and states of the closed-loop system are proved ultimately bounded by Lyapunov stability theory. Simulation results of six-degree-of-freedom show that the tracking speed of control scheme is fast and the robust is strong. The control scheme has high adaptability for uncertain parameters.