Ground Marking Detection During Automatic Landing Of UAVs

With the progress of the economy of our country and the rapid development of the society,our comprehensive national strength has steadily risen.With the continuous development of science and technology and the increasing use of UAVs,UAVs have been widely used not only in the military field but also in the transportation field.UAVs are increasingly close to people's daily lives.The UAV is applied to the emergency lane of the expressway management system.Using the UVA with a camera for road cruise can greatly facilitate the daily management of the expressway,reduce the potential safety hazard and reduce the consumption of resources such as manpower and material resources.However,due to the large power of UAVs and the poor battery endurance,UAVs need to land at specific place in time for charging.This paper focuses on the issue of automatic landing at specific place of UAVs.The main work is as follows:(1)Designing a new landmark which consists with a circle and a rectangle.Using color image segmentation,edge detection and tellipse fitting to judge whether it is a target landing site in the image.(2)An comer detection method is experimented in this paper.And we study the camera calibration algorithm proposed by Zhang Zhengyou.Then doing camera calibration and calculate the camera intrinsic parameters.(3)When it is recognized that there is a target landing site in the current field of view,the UAV hovers about 1 to 2 meters away from the target landing site.As the camera intrinsic parameters are known,a method for calculating the relative distance between the UAV and the landing site and a method of pose estimation are proposed under the background of the project.This paper does research on the feasibility of UAVs landing on the specific area in the system by above three steps.Experimental results indicate that the methods proposed in this paper and the landmark designed in it can get the position of the landing site and it can be accurate to centimetre level.