Research On Fuzzy Adaptive Control Method Based On DSC For Hypersonic Vehicle

Hypersonic vehicle refers to the aircraft whose flight Mach number is greaterthan5. Although hypersonic vehicle has a promising prospect in both military andcommercial applications because of its high flying speed, strong penetration abilityand reusability, the highly nonlinear model, uncertain parameters and disturbances inthe complex flight condition bring great challenges to the design of control system,this problem has become the core of research both at home and abroad gradually.Under this background, considering the hypersonic vehicle models, in order to realizethe accurate tracking of the height/velocity and attitude Angle commands, thisdissertation begins a study in the following three aspects:For the uncertainty result from parameter perturbation of the aircrafts, a fuzzyadaptive nonlinear control approach based on dynamic surface control (DSC) isproposed to design the altitude and velocity tracking controllers respectively. Thefuzzy adaptive method is used to identify the uncertain function terms on-line; andNussbaum gain approach is adopted to restrain the influence caused by the uncertainvirtual control gain and avoid the problem of singularity; the dynamic surfacecontrol(DSC) is applied to avoid the complex derivative. The controlled system areensured to be semi-global uniformly ultimately bounded. The simulation results showthat this proposed approach is feasible and effective for hypersonic vehicle control.For the uncertainty result from parameter perturbation and external disturbanceand the big computing burden caused by the great number of adaptive parameters, anadaptive fuzzy robust control strategy based on ISS and small gain theorem isproposed in this dissertation. The Takagi–Sugeno type (T-S) fuzzy logic systems areused to approximate the total uncertainties in the simplified model, where the adaptivealgorithm is adopted to approximate its boundary of modeling error which is offset inthe control law, however the items related to tracking error z are characterized byanother dynamic, the final stability for the interconnected systems z is analyzedby the use of input-state stability and small-gain stable. The proposed control methodcan not only avoid the singularity phenomenon of the controller, but also minimizethe number of on-line learning parameters to2. For the uncertainty result from parameter perturbation and external disturbancein the angle model durning the reentry phase, an adaptive fuzzy controller based onextended states observer (ESO) is proposed to track the attitude angle commands. Forthe slow loop, the fuzzy logic systems are used to approximate the uncertain terms; onthe other hand, the ESO is designed to observe the uncertain terms in the fast loop,which can alleviate the computational burden. On the basis of the preceding methods,the DSC method is applied to design the controller which would simplify the controlalgorithm dramatically and avoid the phenomenon of ‘explosion of Parameters’. Thecontrolled system are ensured to be semi-global uniformly ultimately boundedaccording to the stability analysis. The simulation results show that this approach canrealize the accurate tracking of the attitude angle commands.