Design And Implementation Of Control System For A Quadrotor

A quadrotor is a kind of electric, multi-rotor vertical take-off and landing, remote control or autonomous aircraft. It has the advantages of simple structure, low cost, landing point and hover. With the development of science and technology, a quadrotor has made great progress and has been widely used in civilian and military fields including aerial photography, forest fire prevention, investigation and rescue. In this thesis, the control system of a quadrotor is designed based on PID control algorithm and Linear Active Disturbance Rejection Control (LADRC) algorithm, respectively.For the control problem of a quadrotor, the present thesis analyzes the working principle, describes the ground coordinate system and body coordinate system, and establishes mathematical model based on Newton-Euler equations. Then, the attitude angles of a quadrotor is represented by a quaternion and the inertial measurement unit module is designed to estimate the attitude based on a complementary filter algorithm. The hardware structure and the control system structure of a Smart-QR quadrotor are introduced.First of all, a cascade PID control algorithm is used for the quadrotor. The control system is divided into four dependent channels. The outer loop and inner loop, respectively, are employed to control the attitude angle and the angular velocity. Such algorithms are validated by two steps. In the first step, the control system is built in Matlab/simulink, and the simulink experiments verify the effectiveness of PID algorithms. In the second step, the program is written in Keil MDK and downloaded to Smart-QR’s microprocessor, and the actual experiment is implemented for displaying the effectiveness of the proposed algorithm.Secondly, the LADRC is adopted for a quadrotor because the Smart-QR may mount some equipments for the practical requirements which results in re-tuning the PID parameters. In order to avoid a large error, a tracking differentiator is designed to arrange the trasnsient dynamics of an input signal. A linear extended state observer is designed to estimate the system’s overall disturbance and compensate such that the system is reduced to a unit gain double integrator, which is easily controlled. The simulation of such a closed-loop system is established in Matlab/simulink and the real experiment is provided to verify the effectiveness of an LADRC algorithm.