| Autonomous flight ability of quadrotor aircraft is the premise of autonomous intelligent operation.The ability of trajectory planning is an important basis for autonomous flight of aircraft.In the actual flight process,the UAV needs to deal with the unknown obstacles and adjust the flight path.The trajectory optimization problem(TOP)becomes a large-scale optimization problem due to the strong nonlinear model and multiple complex constraints of the quadrotor aircraft.As a result,the traditional trajectory optimization algorithm is not accurate enough or even difficult to solve the optimization problem.In order to improve the solution accuracy of the quadrotor aircraft trajectory optimization problem and the autonomous flight efficiency of the aircraft,an improved trajectory optimization algorithm for a quadrotor aircraft based on the Gaussian pseudospectral method is proposed in this thesis,and an obstacle model is constructed for obstacle avoidance so as to achieve the off-line trajectory planning.Meanwhile,by combining with the Active disturbance rejection control algorithm,the planning trajectory tracking control research is carried out.The research work of this thesis is summarized as follows:(1)The trajectory optimization model of quadrotor aircraft is established.By analyzing the structure and working principle of the aircraft,the dynamics equation of the system is established.The general optimal control problem is transformed into a standard Bolza form problem suitable for pseudo-spectral method,which provides a model basis for the study of trajectory optimization of quadrotor aircraft.(2)Aiming at the problem of insufficient accuracy in solving the trajectory optimization of complex systems with multiple constraints,an improved Gaussian pseudospectral method(GPM)smoothing penalty(GPM-S-P)algorithm is proposed in this thesis by combining the GPM method(one of the direct methods),smoothing function and penalty funciont.For trajectory optimization without path constraints,the terminal constraint processing method is introduced,and the terminal constraint processing method is judged by comparing the terminal constraint conditions.For the trajectory optimization problem with path constraints,the inequality constraints are equivalent and smoothed by Max function.At the same time,a new state variable is added to the objective function.Finally,the simulation test results show that GPM-S-P improves the solving accuracy and greatly improves the flight efficiency compared with the traditional trajectory optimization algorithm.The effectiveness of the algorithm in the obstacle avoidance problem of trajectory optimization of quadrotor aircraft is verified.(3)Optimization and tracking control are designed for aircraft trajectory.Firstly,a double closed-loop PID controller based on position and attitude was designed to track the trajectory optimization results,and the dynamic characteristics and tracking errors of the control system were analyzed.In order to improve the control defects of traditional proportional integral differential control,this thesis designed a trajectory tracking control algorithm based on GPM-S-P combined with ADRC.Compared with the trajectory tracking results under PID control,the trajectory tracking results under ADRC control are closer to the desired trajectory,which verifies the high-precision tracking control ability of the proposed control algorithm for the desired trajectory.(4)Finally,based on the simulation verification of the trajectory optimization algorithm,a real aircraft verification test is carried out based on the PX4 flight control system.The hardware and software experiment platform is built.And the specific flight experiment scheme and trajectory control program flow are designed.The position information of aircraft is measured by positioning system and barometer,and the result of obstacle-avoidance flight test is analyzed.The experimental data show the effectiveness and reliability of obstacle-avoidance flight. |
