Research On Fuzzy Sliding Mode Control For Hypersonic UnmannedAerial Vehicle

Hypersonic unmanned aerial vehicle (HUAV) is labeled as the flag of the leap-forwarddevelopment of weaponry. Its complex dynamic greatly affects how the aircraft is flying and bringstremendous challenges and requirements to control it. In this thesis，three relative problem, i.e.modeling and analysis of conceptual HUAV, nonaffine system control and fuzzy sliding mode controldesign, are studied.Firstly, based on the reference both at home and abroad, a simulation longitude model ofconceptual HUAV is presented. Different from previous literature, the relationships between thecoefficients and control surface deflections are represented by nonlinear functions, therefore theequations contained in the established mathematical model imply the control input, thus the wholesystem represented by that mathematical model is nonaffine nonlinear. Simulation results showed thatits open-loop characteristics are consistent with the existing conclusion, therefore this model has acertain level of representativeness so that our subsequent studies can base on the establishedmathematical model..Secondly, to develop our ensuing research step by step, the hypersonic unmanned aerial vehiclelongitudinal model is simplified to conform to the form of a class of single-input and single-outputnonaffine nonlinear system. Then, for the formatted system, an adaptive fuzzy control using linearsliding surface is designed so that the system output can track the desired signal asymptotically. In thedesign of the controller, fuzzy logic system is used to approximate the unknown functions. In order toavoid parameter drifts,-modification is introduced in the fuzzy parameter adaptive law. Theuncertainties are only required bounded and yet their supremums or infimums is not needed. Toenhance the robustness of the system, robust control part is added to compensate the approximationerror and disturbances. Finally, by constructing reasonable Lyapunov function, the stability of thewhole closed-loop control system is testified. Through the simulation of simplified HUAVlongitudinal controlled system, the effectiveness of the proposed method is verified.Then, the unreduced HUAV longitudinal model is transferred as a typical multiple-inputmultiple-output nonaffine system with equal inputs and outputs for the controller design. In order togain better dynamic performance and steady state performance at the same time, instead of linearsliding surface, a new nonlinear integral sliding surface is used to design the fuzzy nonlinear integralsliding mode controller. In addition, to avoid arguments explosion, the control system only tunes theestimations of the norm of unknown parameters online, which effectively reduces the response time.Lyapunov function of the whole system is designed to prove that all the signals in the closed-loopsystem are bounded and the tracking errors converge to zero asymptotically. Apply the proposedmethod to the HUAV model for simulation, and the simulation results demonstrate the effectivenessand robustness of the controller.Finally, the rapid time-varying volatility of HUAV requires the system states to converge to thedesired trajectories in a finite time. For this purpose, the fuzzy fast terminal sliding mode control arepresented for HUAV multiple-input multiple-output system. Without loss of generality, the stability ofthe whole closed-loop is proved via Lyapunov method. In the end, simulation results show the systemcan track the desired output signal quickly, i.e. the effectiveness and robustness of the proposed method is verified.