Unmanned aerial vehicle autonomous guided landing method based on monocular vision navigation

• Main Authors: HAN YUQI, DU ZHONGQIANG, LIU SHAOPENG, TANG LINBO, XU JUN, TAN QIAOZHI
• Format: Patent
• Language: Chinese; English
• Published: 06.12.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The invention discloses an unmanned aerial vehicle autonomous guided landing method based on monocular vision navigation, and belongs to the technical field of unmanned aerial vehicle landing control systems. Identifying an airport from the obtained image through a YOLOv3 model, and cutting an area containing an airport runway according to a detection frame; segmenting an airport runway through a sepnet model, fitting a runway edge through a mask straight line of the segmented airport runway, and obtaining pixel coordinates of the vertex of the runway; and calculating the positioning coordinates of the unmanned aerial vehicle by combining the latitude and longitude information of the runway vertex, the attitude angle of the unmanned aerial vehicle and the like. And finally guiding the unmanned aerial vehicle to correct the course to complete landing. The method solves the problems that the autonomous landing precision of the unmanned aerial vehicle is low and the method universality is poor only depending on