基于阶层标识的无人机自主精准降落系统

随着微小型无人机（UAV）在航拍、测绘、环境监测、快递投送等民用领域的广泛应用，对微小型无人机的可用性和可靠性提出了更高的要求。为了使微小型无人机能够精确地完成自主降落，由于计算机视觉部署成本低、独立性强、信息丰富等特点，提出了通过识别匹配一种多层嵌套二维编码的阶层降落标识来进行相对定位的算法，并展示了与之对应的阶层标识检测及定位的无人机自主降落系统，由于编码的信息量与其他系统相比，具有高低空的高识别率、编码空间大等特点，故此系统可同时支持单个或多个停机坪的配置，且成本低廉，无需添加机载设备成本。最后对该系统进行仿真验证和实飞测试，表明所提出算法能够有效地实现无人机全自主降落。

Multi-level marker based autonomous landing system for UAVs

With the wide application of micro Unmanned Aerial Vehicle (UAV) in civil fields such as aerial photography, mapping, environmental monitoring and courier delivery, there is a higher requirement for the availability and reliability of micro UAV. In order to make the micro UAV complete autonomous landing accurately, based on computer vision which has the advantages of low cost, strong independence, and abundant information, We present a relative localization algorithm that identified and matched a multi-leval nested Marker,and show the corresponding autonomous landing system of UAVs with multi-leval Marker detection and orientation. Compared with other systems, the amount of information encoded has the characteristics of high recognition rate and large coding space. Therefore, The system can simultaneously support the configuration of single or multiple apron platforms, and low cost, and without add airborne equipment cost. Finally, the simulation verification and real flight test of the system show that the proposed algorithm can realize the autonomous landing of UAVs.