Research On Trajectory Tracking Control Of Fixed-wing Unmanned Aerial Vehicle

The fixed-wing UAVs are playing more and more important roles in modern military fields.They have gradually become significant research contents because of their strong flight endurance,fast cruising speed and high cost performance.Trajectory tracking control of UAV,whose effects will directly affect flight security and mission result,is the foundation of cruising task and target tracking.This paper presents the trajectory tracking control problem of a fixed-wing UAV.As we know that fixed-wing UAV has characteristic of nonlinear,under-actuation and strong coupling as well as sensitive to disturbance and complex flight environment.All of this increase the difficulty of trajectory tracking control.The control system of trajectory tracking consists of three segments: navigation,guidance and control.The main procedure of navigation is planning flight trajectory according to flight environment and optimization performance index.Guidance means that UAV's flight states determine the desired attitude and velocity.Finally,the UAV will regulate thrust and deflection to realize low-level control.According to the analysis of system model and attitude description with quaternion,this paper proposes a back-stepping sliding mode method based on extended state observer(ESO)to depress the effect of external disturbance and internal model uncertainties.At last,UAV complete the optimization of flight trajectory through particle swarm optimization algorithm.In this thesis,my main works include:(1)Firstly,the system model is established with quaternion as the fixed-wing UAV attitude parameter.This kind of modeling can avoid the singularity phenomenon and complicated triangular calculation when using Euler angles as parameter of attitudes.The whole system can be divided into translational and rotational motion equations.(2)Based on the model of fixed-wing UAV,attitude and velocity controllers are designed.Considering model uncertainty and external disturbance,this thesis proposes a back-stepping sliding mode control method based on extended state observer to depress disturbance and achieve the desired value.(2)Based on the current real position and desired position,we can design a virtual control variable and convert this variable into desired attitude and velocity value of UAV.Then executing differential calculation on desired quaternion by tracking differentiator in order to obtain desired angular velocity and using particle swarm optimization algorithm for planning optimal trajectory in complex environment.