Vision-based Research On Navigation And Landing System Of Quadrotor Unmanned Aerial Vehicle

At present,more and more unmanned aerial vehicles(UAV)are used in military fields and civil fields.With the application of UAV becoming more and more widely,the recovery of UAV has become the focus point of research.UAV flight accidents,mostly produced in recovery process.Therefore develop a safe recovery method has become the most important thing.With the development of computer vision technology,the research on UAV self-recovery and autonomous landing based on the vision has become a hot issue.Under this background,this paper deeply researches and studies about the vision-based autonomous landing and recovery process of Quadrotor Unmanned Aerial Vehicle,and designs the navigation and landing system in Android mobile platform.In view of the ultimate goal of this paper,the main content of the study is showed as follows:(1)In order to extract the landing marker by using the OpenCV visual library based on Android platform,finally,the ORB features point detection and matching algorithm was used.And through the program verification,ORB algorithm can accurately extract the landing marker.(2)In order to control the UAV in Android system,this paper,based on the classical visual control method of the UAV,presents a new method by using the physical characteristics of Android mobile phone.(3)In the end,this paper describes the realization of vision-based navigation and landing system by writing Android mobile application.The application includes the landing marker extraction method based on the ORB algorithm and the vision-based controlled method of the UAV.By analyzing the experimental results,we can draw the conclusion that this vision-based navigation and landing system designed in this paper can be used in reality.