Research On Motion Planning Method For Autonomous Landing Of Quadcopter

Multicopter has been widely used in many fields because of its vertical take-off and landing capability,good maneuverability,strong environmental adaptability and other advantages,but its short endurance limit its further application.Autonomous landing technology can not only effectively solve this problem,but also improve the intelligent level of multi rotor UAV and expand its application range.In practical application,multicopter is often in a complex environment,such as obstacles in the environment.Therefore,the research on autonomous landing technology of multicopter in complex environment is very important.In this paper,the autonomous landing technology of quadcopter is researched based on the method of motion planning.The main contents are as follows:(1)A set of quadcopter software and hardware system and integrated simulation system are built,and on this basis,the algorithm studied in this paper is implemented and verified;(2)Based on the differential flatness property of quadcopter,a nonlinear geometric controller is designed,which can realize the flight control of large maneuver and improve the ability of trajectory tracking;(3)For the autonomous landing of quadcopter on the static platform,the improved rapidly-exploring random graphs algorithm is used for path planning;the trajectory planning model is designed based on optimal control;the yaw angle planning algorithm is designed to solve the problem that the visual sensor FOV limit affects flight safety;The real-time replanning strategy ensures quadcopter to complete the autonomous landing task in the obstacle environment under the premise of ensuring safety;(4)Based on the motion planning method of static platform landing,the strategy of optimal landing trajectory generation is to fit and predict the trajectory of mobile platform,generate multiple trajectories and select the optimal landing trajectory;Taylor expansion and constrained optimization model are used to complete the trajectory fitting and prediction of landing platform,which can realize the state prediction of any unknown mobile platform model;The path planning algorithm,yaw angle trajectory planning algorithm and replanning strategy are improved to improve the performance of the autonomous landing algorithm.In this paper,the performance of each part of algorithms is tested through simulation experiments,and then the trajectory tracking control and motion planning are combined in the three-dimensional real simulation environment to realize the autonomous landing of quadrotor UAV in a complex environment,which further verifies the effectiveness of the autonomous landing method designed in this paper.