GENERAL INSTRUCTIONSUAV (unmanned aerial vehicle) has an important value of application both in military and civil fields for its characteristics such as low risk, good flexibility, high reliability and strong adaptability. However, more and more requirements are increased with the flight control system. A lot of control strategies are put forward to meet multiple performance objectives, such as closed-loop stability, fast transient response and precise tracking. Among them, the sliding-mode control is widely used in flight control system [1][2][3][4].Due to the changes of the flight environment during flight and the inherent uncertainty in flight control system which may cause the decline of flight control system, even the instability of the aircraft, thus, the robustness and efficiency are required. The sliding-mode control provides a robust and effective method of controlling due to its outstanding advantage that the system dynamics will be insensitive to parameter variations and disturbances once the system state reaches a sliding surface [5][6][7].Besides, the actual flight control system is always subject to various physical constraints. In [8],an active fault accommodate strategy is put forward for flight control system in the presence of input constraints. This method improves the tracking performance and eliminates the influences of actuator faults. In order to deal with the input constraints and severe actuator failures, [9] proposes a novel control scheme in which the control allocation method is incorporated with model reference adaptive control, realizing the global stability of the system and the asymptotic state tracking. Apart from input constraints, the flight control system also suffers from state constraints due to the limited input torque, working range and measurement speed. However, there are rare researches on state constraint in flight control system. A two-loop performance-oriented control approach is given for a class of multiple-input-multiple-output (MIMO) system with input saturation and state constraints in [10], getting fast transient response speed as well as good steady-state tracking accuracy without violating constraints. In [11], a global optimal sliding-mode control algorithm is proposed for non-linear system to conquer the acceleration constraint. This algorithm optimizes the performance index. Although these papers mentioned earlier research the state constraint to some degree, they are not combined with the flight control system. As the flight control system is subject to parametric uncertainty and external disturbances, it is hard to make sure that the tracking error converges monotonically as soon as possible without violating state constraints when the initial states are far away from the desired trajectory.Therefore, based on the actual flight control system, this paper presents a strategy of optimal sliding-mode control for non-linear flight control system which is subject to parametric uncertainty, external disturbances, and state constraints which are respectively t...