Design And Implementation Of Rotor UAV Grab System

In view of the high complexity and poor applicability of the current trajectory planning of the rotor flying robot arm,the parallax calculation of the binocular camera to identify the positioning target is very costly in terms of processor resources.In this paper,a monocular camera is used.On the basis of the visually assisted positioning of the monocular camera,the B-spline trajectory planning method is used to realize the fast planning of the trajectory of the UAV.The April Tag visual positioning algorithm is used to ensure the rapid recognition and positioning of the target object by the monocular camera.The visual servo controller based on pose is designed to realize the autonomous grasping of the flying robot arm.This article first designs the flight control system board with STM32F103VET6 as the core processing chip and a three-degree-of-freedom manipulator using Solid Works software.The rotary-wing UAV uses a camera for auxiliary positioning and combines the set points with a cubic B-spline function to generate a smooth flight trajectory.Then,the D-H parameter method is used to complete the modeling of the manipulator,and the forward and reverse kinematics of the manipulator are solved and analyzed.The linear arm interpolation and the circular interpolation method are used to plan the trajectory of the manipulator.After analyzing the camera model,performing camera calibration,hand-eye calibration and citing April Tag visual positioning algorithm for object pose estimation,a posture-based visual servo controller with eyes outside the hand is designed.Finally,the hardware platform of the rotor UAV grab system is completed.The autonomous grab experiment is carried out to verify the stability of the UAV and the feasibility of visual servo control.