Model Predictive Control Of Maneuvering Flight For Unmanned Aerial Vehicles

Birds in nature can decelerate from flat flight rapidly and perch on the branch or other target position.If fixed-wing UAVs can imitate the land mode of birds which means increasing attack angle from flat flight to stall condition and then decelerating rapidly and finally landing on the specified position accurately,no runway landing can be realized.As a result,the application scenarios of fixed-wing UAVs may be expanded.This landing mode mentioned above is called perching maneuver.This paper is mainly pointed at perching maneuver of fixed-wing UAVs through model predictive control.First of all,the piecewise linear model of the perching process of the aircraft has been established.The optimized reference trajectory of perching process has been designed in order to invert the control problem of maneuvering flight for fixed-wing UAV into trajectory tracking problem by linearing the nonlinear dynamic system along the referred trajectory.In terms of nominal piecewise linear model of perching process,the predictive control method based on terminal constraint set has been applied.The switching system has been used to equal piecewise linear model and design predictive controller based on terminal equality constraints.To solve the conservation weakness of the controller,the terminal constraint has been relaxed to terminal constraint set and the controller based on terminal constraint set has been designed.The stability of the whole switching system is also analyzed in the paper.According to the perching process under external wind disturbance,the robust predictive control method has been studied.The tensor product model of aircraft in perching maneuver has been established to transform the linear varying parameter model into convex hull form.Then the infinite time domainmin-max MPC method which can ensure the robust stability has been applied to obtain the state feedback control law and finally the robust predictive control has been realized.Considering the perching process under both external wind disturbance and great initial error,the improved algorithm of robust predictive control has been used.After extracting the limitations of robust predictive control,the robust predictive controllers based on multistep control set and calculation reduce have been designed respectively to deal with conservation,feasible region and calculation problems.