Research On Robust Nonlinear Control For Small-scale Unmanned Helicopter

Compared with other aircrafts, small-scale unmanned helicopters have the advantages ofvertical take-off and landing, hovering, coordinated turn, small size, small weight, low cost,strong concealment and so on, which makes it have widely potential applications in both mil-itary and civil fields. However, the small-scale unmanned helicopters are under-actuated,strongly coupled, unstable and time-varying high order nonlinear systems, which makes theautonomous flight controllers’ design with great challenges still. Therefore, study on the au-tonomous flight controller design for the helicopters is of great scientific significance.The dissertation mainly discusses the model simplification and robust nonlinear control-ler design based on theoretical analysis and numerical simulation.Firstly, this dissertation establishes for a helicopter a fourteenth-order nonlinear dynamicmodel, which is simplified into a linear parameter varying (LPV) model and a nonlinear mod-el augmented with lumped disturbances to meet the need of the controller design.Then, since the shortcomings of linear control methods, a local H_∞optical LPV veloc-ity (LHOV) control method is proposed for the small-scale unmanned helicopter based on thesimplified LPV model.After that, to overcome poor trajectory tracking performance of the LHOV control me-thod, an adaptive backstepping-based flight controller is designed based on the simplifiednonlinear model with lumped disturbances. Furthermore, a robust integral filtering backstep-ping method based flight controller is presented to solve the problem of the differential explo-sion in the adaptive backstepping method.Lastly, to reduce the probability of amplifying the high-frequency measurement noises inthe robust integral filtering backstepping method for the overuse of filters, for the helicopterthis dissertation proposes a robust nonlinear controller with inner-outer loop structure tocompensate disturbances. This controller not only reduces the number of filters to weaken theeffect of amplifying the high-frequency measurement noises, but also reduces calculationworks, which facilitates the control parameter adjustment.In general, the main contents and primary innovations of this dissertation can be summa-rized as follows:1. the background of the small-scale unmanned helicopter both domestic and oversea isintroduced, then an emphasis introduction to the development status of the flight control ispresented.2. A very specific analysis for the dynamic characteristics of the small-scale unmanned helicopter is given, and a fourteenth order full state nonlinear model is established. Then ac-cording to the need of the control design, the dissertation simplifies it to a simpler LPV modeland a simpler nonlinear model which is augmented with lumped disturbances.3. A LHOV control method is presented for horizontal flight of the small-scale un-manned helicopter. This method not only ensures the robust optimization with the function oflocal H_∞optimal performance, but also satisfies the large range robust stability while theclosed-loop system flights to any fixed velocity flight domain, or switches from one fixed ve-locity flight point to others.4. Since the LHOV control method has the shortcoming of large trajectory tracking error,a newly adaptive backstepping method is presented to deal with this problem. this methoduses the adaptive law which is used to estimate the disturbances for compensation in back-stepping controller. Therefore, it can improve the robustness and trajectory tracking perfor-mance. Moreover, it is also proved in Chapter4that the closed-loop system is robustly un-iformly ultimately bounded.5. Considering the large calculation works that the adaptive backstepping method has, arobust integral filtering backstepping controller is proposed in Chapter5. This controller usescommand filters to compute the time derivatives of the virtual controllers to reduce the calcu-lation works. Moreover, it improves the capability of disturbance rejection and removes thestable-state error of the tracking trajectory by adding an integral term and a discontinuous ro-bust term. evidently, it can obtain very good trajectory tracking performance.6. A robust nonlinear control method based on the disturbance compensation is proposed.This method uses dynamic surface control technique to solve the large calculation works; andapplies the inner-outer structure to reduce the use frequencies of the filters to solve the prob-lem of amplifying the high-frequency measurement noises; uses the acceleration measurementvalue to reconstruct the external force disturbances of the X and Y direction in body-fixedframe, and the nonlinear disturbance observer to observe the moment disturbance and errorintegration to compensate the disturbance of the Z direction in the body-fixed frame. As a re-sult, it obtains better robustness and disturbance rejection capability than adaptive backstep-ping method and robust integral filtering backstepping method. At last, some problems are proposed for further research and exploration after the sum-mary of this dissertation.