Research On Control Technology Of Quadrotor Unmanned Aerial Vehicle

The quadrotor UAV is a kind of6Degree of Freedom Vertical Take-Off and LandingUAV, which can fulfill missions that fixed-wing aerial vehicles cannot achieve, such ashovering, flying with low speed, vertical take-off and landing and flying indoor. Comparingwith a traditional helicopter, it has the advantages of simple structure and control, lowrequirement of manufacturing accuracy, strong stability and weak gyroscopic effects. Inrecent years, the quadrotor attracts much attention due to its applications in unmannedinspection, traffic supervision, forest fireproofing, city patrolling, and becomes a hot researchobject.Due to the dynamic characteristics of multivariate, coupled and underactuated, thecontroller design of the quadrotor has always been a key point. The work of this paper mainlyfocuses on the flying control problem. Various simulations and experiments are carried out toanalyze and validate the control schemes proposed in this paper. The main work andcontributions are as follows:1. Nonlinear modeling and dynamic analysis of the quadrotor UAV. The traditionalmodel of the quadrotor UAV uses the Euler angles or the rotation matrix as system states todescribe the attitude dynamics of the aircraft. The Euler angle method has a singularityproblem and it’s needed to solve transcend functions; on the other hand the rotation matrixmethod has constraint redundancy. These problems increase the difficulty of controller design.In view of the above problems, an attitude model is established based on the quaternion andthe Modified Rodrigues Parameters. A system error model is also built for the task oftrajectory tracking control. The flow field and strength analysis are conducted based on thedynamic models. In addition, the wind resistance ability and structural strength are analyzed,which forms the basis of structure design and component selection.2. System design and implementation of the quadrotor UAV. In terms of the dynamiccharacteristics and the analysis results of structural strength, a hardware system of thequadrotor UAV is designed based on an embedded system of DSP/FPGA. A type of BLDCMis chosen as the executive element. The back-EMF zero-crossing detection principle is used todetect the position of the rotor in real-time. A control program of the motor system is designedas well. A quaternion-based direct navigation equation is established and an UnscentedKalman Filter (UKF) is used to achieve the online estimation of the flying parameters of thequadrotor UAV. 3. Position controller design based on the simplified model of the quadrotor UAV. Thedynamics of the quadrotor UAV can be simplified when the Euler angles are close to zero.Therefore, on the basis of simplification, the traditional PID method and sliding mode controlmethod are used. Simulations are implemented to analyze the control effect.4. Trajectory tracking control based on the nonlinear system model. In order to controlthe6-DOF quadrotor aircraft in the presence of constant parameter uncertainties and unknowndisturbance, a model reference adaptive backstepping control method and an adaptiveterminal sliding mode control method are proposed in this thesis. Due to the high-ordercharacteristic of the quadrotor UAV, the dynamic model is decoupled into a positionsubsystem and an attitude subsystem, so as to decrease the dimension of the system. Thethought of cascade system analysis and properties of rotation rigid body are applied to analyzethe system stability.5. Research on the experiments of the quadrotor UAV. Several key problems areexpounded during the system establishment. The experiments of the control schemesproposed in this paper are implemented to validate the control effect based on the quadrotortestbed. The performances are analyzed and conclusions are received based on the compare ofthe control effects.