Research On Design And Realization Of Control System For A Quadrotor Unmanned Aerial Vehicle

Quadrotor unmanned aerial vehicle(UAV), as one kind of the multi-rotoraircrafts, processed great potentials in both civil and miltary applications because ofits excellent characteristics such as small volume, simple structure, flexible operation,vertical takeoff and landing, etc. The design, manufacture, debugging and realizationof a quadrotor in this thesis, are based on the present research progress in this field,and this research is funded by open issued from Shenyang Institute of AutomationCAS(Research on indoor3-D enviromental understanding and autonomous control).First, a quadrotor UAV platform was established for further research includingattitude and heading measurement and flight control.The Newtown Principle wasemployed to derive the dynamic equation of the system according to the flightprinciple of a quadrotor in which the precisely measured parameters were evolved.The entire UAV system was composed by two core parts, i.e. the Attitude&Heading Reference System(AHRS) and the flight control system. Attitude, positionand velocity signals feedback to control system was provided by AHRS, in which dataacquisition from all sensors was conducted. In the flight control system, themicroprocessor generates the control signals in the form of PWM to drive four motorsaccording to the difference between the control commands from the remote controllerand feedback signals from the AHRS.Two PCB boards encompassing STM32ARM Cortex-M4core based highperformance microprocessor, MEMS IC and other ICs were deviced andmanufactured after numerous debugging and validation test. One board noted asNavigation Board is used for the AHRS while the other one which is used by theflight control system is called Flight Control Board. Hierarchical and structuredsoftware was developed after the design and implementation of hardware. Thesoftware system consists of three different but correlated layers: the underlyingdriving layer, the functional modules layer and the application layer.An extended Kalman filter(EFK) was introduced in the AHRS to posture optimalattitude estimation of the quadrotor UAV for the flight control system by taking theadvantages of MEMS three-axises gyroscope, accelerometer and magnetometer intoconsideration. All the raw data were sent to the EKF after front-end drift-correctionand low-pass digital filtering to avoid high frequency disturbance. The attitude andposition controller was designed based on the dynamic model previously established.Numerical simulation was carried out in the Matlab/Simulink, which showed theeffectiveness of the control algorithm.Finally, the proposed attitude estimation algorithm and the flight controlalgorithm was programmed in the circuit boards for experimental test and flight validation. Experimental result showed the accuracy of the attitude measurement androbustness of the attitude and position controllers.