Trajectory Optimization And Attitude Tracking Control For Reusable Launch Vehicle

For the requirement of "rapidity,maneuverability,reliability and low-cost" for space launch system,Reusable Launch Vehicle(RLV)has become a new type of aerospace vehicle,which can fly between earth and orbit with reusability.It owns the ability of rapid deploy,reconstruction,expansion and low-cost of space transportation.However,on account of the characteristics of large-airspace and cross-speed of RLV,the speed,altitude and attitude change largely during the reentry phase.The model,which is influenced by external disturbance and parameter uncertainties,reflects serious nonlinearity and coupling characteristic.All of these factors make the design of RLV trajectory and control system become a very important and challenging task in the current field of aeronautics and astronautics.In this dissertation,RLV trajectory optimization and attitude control are mainly studied for the purpose of a safe and stable reentry flight.Firstly,considering the complex path constraints and characteristic in Terminal Area Energy Management(TAEM)phase,the adaptive Gauss pseudospectral method(GPM)combined with interval analysis method is applied into RLV trajectory optimization for the first time and solves the problem of complicated three-dimensional(3D)trajectory optimization with superior rapidity.In this strategy,TAEM trajectory design is transformed into a multi-stage complex trajectory optimization problem by setting up waypoints on the projection of the heading alignment cylinder(HAC),and meanwhile the continuity condition is proposed to guarantee the continuity of two adjacent subphases.The adaptive GPM can improve the accuracy and the interval analysis can increase the rapidity of the trajectory optimization.Numerical simulation verified the feasibility of the designed trajectory.Secondly,considering the influence of additive model uncertainties and external disturbances during reentry phase,an integrated finite-time observer and controller design scheme is proposed to solve the RLV's robust and exact attitude control.Based on single variable and multivariable super-twisting algorithm,respectively,two novel disturbance observers are proposed in this dissertation.And then RLV attitude dynamic model is divided into two subsystems based on the multi-time scale theory.For each subsystem,a smooth second-order sliding mode controller and a multivariable super-twisting sliding mode controller are designed on the basis of the finite-time disturbance observer,respectively.Two kind of integrated finite-time observer and controller design schemes both can realize the RLV finite-time attitude tracking and reduce the control chattering effectively.The simulation results show that both two control strategies have perfect tracking performance.Thirdly,considering the influence of multiplicative torque perturbation,model parameter uncertainty and external disturbance with unknown upper bounds during reentry phase,two novel kinds of adaptive multivariable super-twisting control methods are proposed and combined for RLV based on double-loop control structure.The finite-time multivariable sliding mode control strategy with gain adaption can solve the problem that the control gain is too large for over-estimation of the upper bound of disturbance,reduce the control chattering and guarantee that RLV can track the reentry attitude control command accurately in finite time.The simulation results show that adaptive multivariable super-twisting control strategy can overcome multiplicative torque disturbance,and improve the control precision while the robust performance is not influenced.Finally,an adaptive observer-based finite-time continuous fault tolerant control strategy is proposed to deal with the actuator fault during the reentry phase.Firstly,a second-order fault dynamics modeling is studied when considering the actuator effectiveness loss fault and a novel gain-adaptive multivariable observer is presented in this dissertation in order to estimate the fault and disturbance uncertainty.On this basis,combining the terminal sliding mode and super-twisting sliding mode,a finite-time second-order sliding mode controller is proposed to to realize the finite-time attitude fault tolerant tracking and control with adaptive observer.In addition,the problem of unmeasured state derivative can be addressed by using the high-order differentiator.Finally,simulation results of X-33 are provided to verify the effectiveness and robustness of the proposed fault-tolerant controller in tracking the guidance commands.