Dynamic Sliding-Mode Control Scheme for Hypersonic Vehicles Considering Nonminimum Phase Characteristics and Performance Recovery

Abstract

This paper presents a performance recoverable control scheme for air-breathing hypersonic vehicles (HSVs) with nonminimum phase characteristics. As elevator and throttle are the only two control inputs available for longitudinal trajectory tracking of HSVs, there are two representative problems emerged. One is the nonminimum phase behavior in pitch dynamics of HSVs due to elevator-to-lift coupling, which prevents the application of standard inversion-based control techniques. The other is the engine performance recovery control with the disabled propulsion system, when flight states of HSVs exceed the safe working boundary, causing engine flameout. In view of the above, a comprehensive control scheme is proposed to realize performance recoverable and high-precision trajectory tracking of nonminimum phase HSVs. First, a dynamic integral sliding-mode (DISM) control method based on the Byrnes-Isidori (B-I) normalized form is proposed, achieving asymptotic tracking of velocity and flight path angle (FPA) while stabilizing internal dynamics. The control method transforms the FPA tracking problem into a stabilization problem of an augmented system consisting of internal dynamics and dynamic compensator, making closed-loop pole adjustable, and thus improves the tracking performance. Second, on this basis, a performance recoverable control scheme is proposed to achieve the automatic performance recovery of HSVs when the engine flameout occurs due to the stall of the HSVs. It achieves temporary acceleration by adjusting the trajectory, thus obtaining the conditions for engine restart. The simulations of the nominal condition and engine flameout condition are presented, respectively, to verify the proposed control scheme. From the simulation results, the proposed control scheme is shown to have superior tracking accuracy with internal dynamics stable, and the robustness is fully verified by the Monte Carlo simulations. Especially, speed autonomous recovery and engine restart are achieved during the engine flameout, which demonstrates the effectiveness of the proposed performance recovery control scheme.