Research Of Autonomous Landing Control Algorithm For Quadrotor Based On Fuzzy Prediction

Quadcopter has the characteristics of non-linear,strong coupling and easy to be disturbed.Therefore,it is one of the key and hot issues in the research and application fields for quadcopter to realize the autonomous landing system with high precision.For the autonomous landing of quadcopter,the accuracy of position estimation is the premise of the accurate landing of quadcopter.Meanwhile,it is difficult to design the control system of the quadcopter in the landing process because of the disturbance of the air flow on the ground.Therefore,the navigation and control system for autonomous landing of quadcopter is studied in this paper.The research of this thesis is as follows.Firstly,aiming at the problem of signal missing in traditional GPS method,an integrated navigation based on vision and INS is designed to computer the position of quadcopter.Meanwhile the mismatch problem of data update period between vision navigation and INS in integrated navigation is solved by fuzzy prediction.Fuzzy prediction is used to predict the visual observation data synchronously when the Kalman state equation is updated.Which reduces the error caused by the direct fusion of visual navigation and INS data.The simulation results show that the integrated navigation algorithm designed in this thesis has a higher position estimation accuracy than the conventional Kalman filter.Secondly,a cascade altitude position-velocity control system for quadcopter based on PID and linear active disturbance rejection control(LADRC)is designed to deal with the disturbance of air flow on the ground during autonomous landing process.Altitude position is the main variable since altitude velocity is the secondary variable.The altitude position is tracked by PID control.Meanwhile,the disturbance can be estimated in real time by the extended state observer of LADRC.The simulation results show that the altitude control system designed in this thesis has shorter anti-interference regulation time than the conventional algorithm,which improves the anti-interference of quadcopter's altitude control.Finally,the hardware platform of autonomous landing system for quadcopter based on vision and inertial navigation system(INS)is built in this thesis.Therefore,the integrated navigation algorithm and autonomous landing control system for quadcopterdesigned in this thesis are verified by airborne experiment.The experimental results verify the effectiveness of the autonomous landing control algorithm for quadcopter based on fuzzy prediction in this thesis.