Research And Design Of The Control System Of A Miniature Quadrotor UAV

A Quadrotor Unmanned Aerial Vehicle(QUAV)generally refers to an aircraft having a symmetrical structure of four rotors and controlling its flight with a radio remote controller or its own program.Since entering the 21 st century,the QUAV has gradually been applied to express logistics,post-disaster search and rescue,film and television shooting,environmental protection,kinds of inspection work and other various fields due to its characteristics such as strong load carrying capacity,easy control,continuous hovering,vertical lifting,and convenient carrying.With the increasing demand for unmanned aerial vehicle performance from all walks of life and it is a typical multi-coupled,non-linear,multi-input and multi-output underactuated complex control system in itself,which led to an upsurge of research in the entire world.The content of this article mainly covers the following aspects:First,the flight control principle of the QUAV was analyzed and a mathematical model was established on this basis,the major ones include: kinematics model based on Euler angles,rotation matrix and quaternion;dynamic model based on position and attitude;control distribution model based on horizontal channel,height channel and attitude channel.Second,in order to understand the flight control system of the QUAV more effectively,the hardware and software system of the QUAV was designed.The hardware system mainly includes: a control module,a measurement and detection module,a wireless communication module,and a power supply module.The software system mainly includes: software development environment,transplantation of the real-time operating system FreeRTOS,and the bottom driver of the measurement and detection unit.Third,in order to obtain the real-time position of QUAV accurately and quickly,this paper based on the establishment of a good linear and horizontal channel linearization model and the selected measurement and detection module,we study the high fusion algorithm based on distributed Kalman filter and the horizontal positionfusion algorithm based on BP neural network.The simulation experiments show that the high fusion effect based on distributed Kalman filter algorithm and the horizontal position fusion effect based on BP neural network algorithm can improve the real-time and accuracy of position information acquisition to a certain extent.Finally,based on the established dynamic model of position channel and attitude channel,the QUAV control system based on PID algorithm and linear auto disturbance rejection control(LADRC)algorithm was studied separately.The experimental results obtained on the Matlab-based QUAV flight simulation platform show that the position and attitude control effects of the QUAV control system based on the LADRC algorithm are obviously better than the PID algorithm.