Research On Position Tracking Control System Of Quadrotor Aircraft

Since the maneuverability of quadrotor aircraft is much better than that of other traditional aircraft in small space,it is often used in underground,indoor,mountainous areas and other similar environments,but these areas where the satellite signals are blocked can not be positioned,which makes the position tracking control of quadrotor aircraft particularly important.In recent years,with the development of machine vision,vision-based positioning methods have emerged and attracted wide attention.Therefore,the position tracking control method based on vision positioning has become one of the solutions to this problem that quadrotor aircraft can not locate.This thesis combines the visual positioning method with the position tracking algorithm to research the position tracking control of quadrotor aircraft.The main work is as follows:Firstly,the structure and flight principle of the quadrotor aircraft is analyzed,the quadrotor aircraft is regarded as a rigid body and the influence factors such as air resistance are neglected.The simplified mathematical model of quadrotor aircraft is established according to the physics principle and the Newton Euler equation.Then,the Co-simulation model of quadrotor with major input-output relationships was constructed using ADAMS and MATALAB.Secondly,the position tracking control system of quadrotor aircraft needs accurate positioning information,which includes attitude angle information and location information.In this thesis,the attitude angle information is obtained by fusing the information collected by accelerometer and gyroscope through complementary filter,while the position information is obtained by visual detection.Thirdly,on the basis of relatively accurate positioning,since the Co-simulation model of quadrotor is decoupled,the position tracking control system with attitude loop as the inner loop and position loop as the outer loop is designed.The results of attitude PID control algorithm and attitude sliding mode control algorithm are compared and analyzed,and the stability of sliding mode control is more in line with the requirements.The sliding mode PID position tracking control algorithm with attitude sliding mode controller as the core and position PID controller as the outer loop control is further designed.Through the Co-simulation results,it can be seen that the sliding mode PID position tracking algorithm is superior to the PID position tracking control algorithm.Finally,in order to solve the problem of visual positioning and improve the tracking control effect of the position tracking control algorithm,the FAST feature point detection algorithm combined with the SIFT feature point detection algorithm is improved.It can be seen from the experimental results that the detection time is effectively reduced and the detected feature points are both time-efficient and rotationally invariant,and the feature point matching rate and stability are higher.The speed information calculated by the improved visual matchingalgorithm is compensated to the sliding mode PID position tracking controller for improvement.The simulation experiment proves that the sliding mode PID position tracking control algorithm with speed compensation has faster reaction time and stronger anti-interference ability.