一种面向移动平台的无人机自主降落控制方法  

A Method for Autonomous Landing of Unmanned Aerial Vehicle Based on Mobile Platform

作　　者：张伟[1] 马珺[1] ZHANG;Wei MA Jun(Taiyuan University of Technology,Taiyuan Shanxi 030600,China)

机构地区：[1]太原理工大学,山西太原030600

出　　处：《计算机仿真》2020年第2期92-96,207,共6页Computer Simulation

基　　金：山西省自然科学基金项目(2015011050)。

摘　　要：准确获取无人机的位姿信息是顺利执行无人机自主导航、着陆的首要前提。由于GPS/INS导航系统的局限性和IMU惯性导航系统的误差,提出了一种基于视觉导航的方法。设计了"H"形图像的着陆标志,对机载摄像机采集的实时影像进行图像处理,利用世界坐标系到图像像素坐标系的映射关系得到基于视觉的无人机位姿估计模型,进而解算无人机当前的位姿估计值。上述方法提高了无人机的着陆安全性。仿真结果验证了算法的有效性和位态信息的精确度。The accurate acquisition of UAV's position and attitude information is the first prerequisite for the successful implementation of autonomous navigation and landing.Due to the limitations of the GPS/INS navigation system and the error of the IMU inertial navigation system,a method based on visual navigation is proposed,which includes designing"H"type landing sign,processing the image collected by the airborne camera,and using the relation between the world coordinate system and the image pixel coordinate system to get the vision-based USV's position and attitude estimation value.Matlab simulation results verify the effectiveness of the algorithm and the accuracy of position and pose information and show that the landing safety of the UAV is improves.

关 键 词：无人机 视觉导航 图像处理 位姿估计 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]