Research On Autonomous Landing Technology Of Accompanying UAV

In recent years,Unmanned Aerial Vehicle(UAV)and its related technologies have developed rapidly,and its application field based on social industry is expanding more and more widely and gradually intelligent,and it can replace workers to complete various tasks in various industries.This has led to the problem of UAV with limited endurance and requiring manual synchronous maintenance operations in complex environments.Therefore,it is increasingly important to research and develop the air-ground cooperative autonomous operation system and use the ground autonomous recovery platform as the automatic apron and energy-exchanging device for the accompanying UAV.Based on the air-ground Cooperative Operation System project,the four-rotor UAV with simple structure and flexible operation are selected as the accompanying UAV,and the autonomous landing technology is studied through vision guidance,the design of control and landing scheme.In this thesis,a computer vision-based autonomous landing scheme for the accompanying UAV is proposed.First,the basic structure and working principle of the four-rotor UAV were studied.The coordinate system was transformed according to the Euler angles method of three-axis rotation,and the dynamics model of the accompanying UAV was established.Secondly,a new marker is designed,and the algorithm of hierarchical target recognition and location is designed.The experimental results show that it can improve the detection and recognition rate of the marker.Then,based on the hierarchical target recognition algorithm,a phased autonomous landing scheme is proposed,which realizes the autonomous landing of the accompanying UAV in two parts,based on the theory of Differential Flatness,a polynomial parameter optimization method for autonomous landing trajectory planning of the accompanying UAV is studied,the model predictive control(MPC)algorithm is used to design and optimize the landing trajectory tracking controller,and the PID control of classical control theory is used to design the position adjusting control module,which is convenient for UAV to apply visual guidance and feedback to control loop to land in the set position.In this thesis,the simulation environment and the basic platform are built,and the experimental results show that the target location algorithm and the vision autonomous landing scheme are reasonable and feasible,it effectively improves the position accuracy of autonomous landing of the accompanying UAV,and lays a theoretical foundation for UAV to carry out tasks automatically and recover energy supply autonomously in complex operating environment,has the practical significance and the application value.